Robotics Seminar @ Illinois

Recording link: https://mediaspace.illinois.edu/media/t/1_mf62q8z9/386332862

Abstract:
Industry is placing big bets on “brute forcing” robotic control, but such approaches ignore the centrality of resource constraints in robotics on power, compute, time, data, etc. Towards realizing a true engineering discipline of robotics, my research group has been “exploiting and exploring” robot learning: exploiting to push the limits of what can be achieved with today’s prevalent approaches, and “exploring” better design principles for masterful and minimalist robots in the future. As examples of “exploit”, we have trained quadruped robots to perform circus tricks on yoga balls and robot arms to perform household tasks in entirely unseen scenes with unseen objects. As examples of “explore”, we are studying the sensory requirements of robot learners: what sensors do they need and when do they need them during training and task execution? In this talk, I will highlight these examples and discuss some lessons we have learned in our research towards better-engineered robot learners.

Speaker Bio:
Dinesh Jayaraman is an assistant professor at the University of Pennsylvania’s CIS department and GRASP lab. He leads the Perception, Action, and Learning (Penn PAL) research group, which works at the intersections of computer vision, robotics, and machine learning. Dinesh received his PhD (2017) from UT Austin, before becoming a postdoctoral scholar at UC Berkeley (2017-19). Dinesh’s research has received a Best Paper Award at CORL ’22, a Best Paper Runner-Up Award at ICRA ’18, a Best Application Paper Award at ACCV ’16, the NSF CAREER award ’23, an Amazon Research Award ’21, and been covered in The Economist, TechCrunch, and several other press outlets.

Recording link: https://mediaspace.illinois.edu/media/t/1_p3hx5f7d/386332862

Abstract:
Recent advances in surgical robotics have enabled innovative techniques that reduce patient trauma, shorten hospital stays, and improve both diagnostic accuracy and therapeutic outcomes. However, despite the many benefits of robot-assisted minimally invasive and endoscopic procedures, significant limitations remain, particularly in the dexterity, intelligence, and autonomy of current robotic systems, as well as in the biomechanical design of medical devices and implants.
One of the most pressing challenges is inadequate dexterity, driven by limited access to target anatomy and insufficient control over rigid instruments and implants. These constraints highlight the need for specialized instrumentation, intelligent sensing, and adaptive control paradigms capable of navigating complex anatomical environments. Advancing autonomy in surgical systems also requires more than robotic intelligence alone, it demands synergistic human–robot interaction, where the system can both assist and adapt to the surgeon’s decision-making process in real time.
This talk will introduce our lab’s efforts in advancing the engineering of surgery (or surgineering) to address these challenges. I will present translational research across several clinical applications, including:
•             Spinal fixation using steerable drills and flexible pedicle screws
•             Colorectal cancer screening with vision-based tactile sensors and complementary AI algorithms
•             In vivo bioprinting for treating volumetric muscle loss via robotic delivery systems
By integrating continuum manipulators, stretchable soft sensors, intelligent implants, and semi/autonomous control strategies, our work aims to fundamentally transform the paradigm of minimally invasive and endoscopic interventions, bringing true dexterity and autonomy to surgical robotics.

Speaker Bio:
Dr. Farshid Alambeigi is the Leland Barclay Fellowship in Engineering Associate Professor in the Walker Department of Mechanical Engineering at The University of Texas at Austin, where he has served since August 2025. He is also a core faculty member of Texas Robotics. Dr. Alambeigi earned his Ph.D. in Mechanical Engineering (2019) and M.Sc. in Robotics (2017) from Johns Hopkins University. In 2018, he was awarded the 2019 Siebel Scholarship in recognition of his academic excellence and leadership. He is the recipient of the NIH NIBIB Trailblazer Award (2020) for his work on flexible implants and robotic systems for minimally invasive spinal fixation surgery and the NIH Director’s New Innovator Award (2022) for pioneering in vivo bioprinting surgical robotics for the treatment of volumetric muscle loss. His contributions have also been recognized with the UT Austin Faculty Innovation Award, the Outstanding Research Award by an Assistant Professor, the Walker Scholar Award, and several best paper awards and recognitions. He serves as an Associate Editor for the IEEE Transactions on Robotics (TRO), IEEE Robotics and Automation Letters (RAL), and the IEEE Robotics and Automation Magazine (RAM).
At UT Austin, Dr. Alambeigi directs the Advanced Robotic Technologies for Surgery (ARTS) Lab. In collaboration with the UT Dell Medical School, the ARTS Lab advances the concept of Surgineering, engineering the surgery, by developing dexterous, intelligent robotic systems designed to partner with surgeons. The ultimate goal of this work is to enhance surgical precision, improve clinician performance, and advance patient safety and outcomes.

In-person event in CSL Studio 1232, Recording link: https://mediaspace.illinois.edu/media/t/1_31klklvi/386332862

Abstract:
Achieving intelligent robotic manipulation requires bridging the gap between multimodal perception and interpretable, safe control. This talk presents a unified research journey toward neural-symbolic control, where robots reason and act through the integration of multimodal sensing, differentiable optimization, and data-driven learning. I will begin with ManiFeel, a comprehensive benchmark for visuotactile manipulation, which establishes a scalable framework to study how tactile sensing enhances policy learning under visually degraded or contact-rich conditions. ManiFeel systematically evaluates sensing modalities, tactile representations, and policy architectures, revealing when and how multimodal feedback improves manipulation robustness. Building on these insights, I will introduce LeTac-MPC, a learning-based model predictive control (MPC) framework that couples high-resolution tactile sensing with differentiable MPC. LeTac-MPC enables tactile-reactive grasping across diverse object properties, achieving robust, adaptive control in dynamic and force-interactive environments. Finally, I will present LeTO, a learning constrained visuomotor policy that embeds differentiable trajectory optimization into neural networks, combining the interpretability and safety of model-based control with the adaptability of deep learning. LeTO enables robots to generate smoother, constraint-compliant, and dynamically feasible trajectories, ensuring safer and more reliable interactions in real-world manipulation tasks. Together, these works chart a pathway toward intelligent, physics-grounded, and neural-symbolic manipulation driven by multimodal sensing and optimization-based learning.

Speaker Bio:
Dr. Yu She is an assistant professor at Purdue University Edwardson School of Industrial Engineering. Prior to that, he was a postdoctoral researcher in the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT from 2018 to 2021. He earned his PhD degree in the Department of Mechanical Engineering at the Ohio State University in 2018. His research is at the intersection of mechanism design, tactile sensing, intelligent control, and robot learning. He is a recipient of the ASME Feudenstein Young Investigator Award (2025), Showalter Early Investigator Award (2024), and the Google Research Scholar Award (2022) and multiple best paper recognitions.

In-person event in CSL Studio 1232, Recording link:
https://mediaspace.illinois.edu/media/t/1_0t8tjx3p/386332862
https://mediaspace.illinois.edu/media/t/1_ixwkqv5f/386332862

Abstract:
Talk 1:
Diagnostic and interventional clinical systems require novel combinations of technology to meet clinical needs. In this talk, we’ll explore two medical problems: access to vision screening and access to traumatic brain injury treatment, and how we might approach these problems using novel systems. In the first, a fully automated retinal imaging system is shown to produce screening-quality retinal images that can be used to assess the health of the eye, serving as the skeleton for future eye examination techniques to be built upon. The second is a novel and compact image-guided surgery system designed to improve the workflow for bedside neurosurgical guidance.
Talk 2:
Large language models (LLMs) have recently demonstrated success in generalizing across decision-making tasks including planning, control, and prediction, but their tendency to hallucinate unsafe and undesired outputs poses risks. We argue that detecting such failures is necessary, especially in safety-critical scenarios. Existing methods for black-box models often detect hallucinations by identifying inconsistencies across multiple samples. Many of these approaches typically introduce prompt perturbations like randomizing detail order or generating adversarial inputs, with the intuition that a confident model should produce stable outputs. We first perform a manual case study showing that other forms of perturbations (e.g., adding noise, removing sensor details) cause LLMs to hallucinate in a multi-agent driving environment. We then propose a novel method for efficiently searching the space of prompt perturbations using adaptive stress testing (AST) with Monte-Carlo tree search (MCTS). Our AST formulation enables discovery of scenarios and prompts that cause language models to act with high uncertainty or even crash. By generating MCTS prompt perturbation trees across diverse scenarios, we show through extensive experiments that offline analyses can be used at runtime to automatically generate prompts that influence model uncertainty, and to inform real-time trust assessments of an LLM. We further characterize LLMs deployed as planners in a single-agent lunar lander environment and in a multi-agent robot crowd navigation simulation. Overall, ours is one of the first hallucination intervention algorithms to pave a path towards rigorous characterization of black-box LLM planners.

Speaker Bio:
Speaker 1:
Alexander Smith is a sixth-year MD-PhD student at the Carle Illinois College of Medicine and the Siebel School for Computing and Data Science at the University of Illinois at Urbana-Champaign. He received his B.S. degree with a dual major in Biomedical Engineering and Computer Science from Saint Louis University in 2019. From 2019-2020, he worked as a Researcher at Johns Hopkins University in the Carnegie Center for Surgical Innovation.
Speaker 2:
Neeloy Chakraborty is a fifth-year PhD candidate at the University of Illinois working in the Human-Centered Autonomy Lab. Prior to that, he completed his M.S. in 2023 and B.S. in 2021 at the University of Illinois in electrical and computer engineering. His primary research revolves around developing reliable technologies to enable safer interactions between humans and automation. Throughout his time at Illinois, he has studied modular approaches for instruction-following embodied AI robots, developed real-time video generation models to allow users to remotely control robots under challenging constraints, designed scalable multi-modal algorithms to detect anomalies in multi-agent settings, and identified hallucinations in large language model generations to proactively avoid failures in decision-making. Neeloy has also conducted research through internships at Dolby Labs, Ford, and Brunswick, tackling problems in generative modeling and reinforcement learning.

Recording link: https://mediaspace.illinois.edu/media/t/1_ndqqu022/386332862

Abstract:
Robot safety is a nuanced concept. We commonly equate safety with collision-avoidance, but in complex, real-world environments (i.e., the “open world”) it can be much more: for example, a mobile manipulator should understand when it is not confident about a requested task, that areas roped off by caution tape should never be breached, and that objects should be gently manipulated to prevent breaking or spilling. However, designing robots that have such a nuanced safety understanding—and can reliably generate appropriate actions—is an outstanding challenge.
In this talk, I will describe my group’s work on systematically uniting modern machine learning models (such as large vision-language models and latent world models) with classical formulations of safety in the control literature to generalize safe robot decision-making to increasingly open world interactions. Throughout the talk, I will present experimental instantiations of these ideas in domains like vision-based robotic manipulation and navigation.

Speaker Bio:
Andrea Bajcsy is an Assistant Professor in the Robotics Institute at Carnegie Mellon University where she leads the Interactive and Trustworthy Robotics Lab (Intent Lab). She broadly works at the intersection of robotics, machine learning, control theory, and human-AI interaction. Prior to joining CMU, Andrea received her Ph.D. in Electrical Engineering & Computer Science from University of California, Berkeley in 2022. She is the recipient of the DARPA Young Faculty Award (2025), NSF CAREER Award (2025), Amazon Research Award (2025), Google Research Scholar Award (2024), Rising Stars in EECS Award (2021), Honorable Mention for the T-RO Best Paper Award (2020), NSF Graduate Research Fellowship (2016), and worked at NVIDIA Research for Autonomous Driving.

Recording link: TBA

Abstract:
Marine robots have undergone significant growth, driven by recent advances in artificial intelligence, sensing technologies, and decision-making systems. As demands for ocean exploration, exploitation, and conservation continue to rise, there is an increasing necessity for advanced autonomy in marine robots at both individual and group levels, as well as for frequent and safe deployment of marine robots at an affordable cost. However, the current marine robots available still fall short in meeting the demands of real-world applications, facing challenges such as robust and safe control in complex environments, self-localization and mapping of their environment, and efficient sensing and coordination within a group. My long-term goal is enabling robots to perform tasks autonomously in challenging marine environments by developing innovative robots and advanced algorithms. In this talk, I will highlight my research in three key areas: 1) learning-based control and navigation of 2D surface vehicles in urban waterways, 2) design and coordination of 2D multi-robot systems on the water surface and 3) developing biologically inspired robots and sensors to address challenges faced by 3D underwater robots and robotic swarms. I will also briefly discuss open research problems on these topics.

Speaker Bio:
Dr. Wei Wang is an Assistant Professor in the Department of Mechanical Engineering at the University of Wisconsin–Madison. Prior to this appointment, he was a Research Scientist at the Computer Science and Artificial Intelligence Laboratory (CSAIL) at the Massachusetts Institute of Technology. He earned his Ph.D. in Mechanical Engineering from Peking University and his B.E. in Electrical Engineering from the University of Electronic Science and Technology of China. Dr. Wang has published extensively in top-tier robotics journals and conferences, including Science Robotics, IEEE Transactions on Robotics (TRO), Robotics and Automation Letters (RAL), Bioinspiration & Biomimetics (B&B), ICRA, IROS, and CDC. His research has been widely featured in international media outlets such as Reuters, NBC, CNBC, and MIT News.

Recording link: TBA

Abstract: The integration of robotic systems alongside human agents introduces significant challenges in decision-making, planning, and control. In particular, navigation algorithms for robots must account for interactions with humans to ensure safety and efficiency. In this talk, I will present an overview of four research directions we’ve explored in our work on Interaction-Aware Planning and Decision Making. First, we will explore interactive path planning in environments with dynamic agents, including approaches based on path-speed decomposition and multi-future planning under optimization-based frameworks. Next, I will discuss sampling-based methods for interactive planning, highlighting how neural networks can be incorporated into these frameworks. Third, we will examine strategies for crash mitigation when formal safety guarantees cannot be established. Finally, I will introduce recent efforts on end-to-end predictive planning approaches that aim to unify perception and decision-making.

Speaker Bio: Dr. Tariq is a Senior Research Scientist in the Cooperative Intelligence – Mobility Research group at the Honda Research Institute, US. He received his B.Sc. degree in electrical and electronics engineering from the Middle East Technical University, Ankara, Turkey, in 2018. He received his M.S. and Ph.D. degrees in electrical and computer engineering from the University of Maryland, College Park, MD, USA, in 2022 and 2023, respectively. His research interests include the application of control, optimization, and learning techniques to robotic systems, with a focus on real-time decision-making and motion planning for autonomous vehicles. He received the Best Student Paper Award (1st place) at the IEEE International Conference on Intelligent Transportation Systems (ITSC), 2021.

In-person event in CSL Studio 1232, Recording link: TBA

Abstract: Robots are becoming important to enable people in their work and daily life, such as in assistive and performance enhancing devices, like prosthetics and exoskeletons, and in teleoperating remote agents, to complete tasks in remote or dangerous locations. The communication between people and their robotic devices must be intuitive and responsive in order to be effective. My work focuses on utilizing the sense of touch for human-robot communication, acknowledging the important role of haptic information in navigating our environment. In this talk, I will discuss my work in two areas of haptic technologies. The first is in the design and integration of haptic devices in human-robot systems. This leads to the second, addressing a bottleneck in device saliency by combining methods from psychophysics and contact mechanics. Through my work I aim to create intuitive communication with haptic systems to create more immersive and efficient interactions for the human utilizing robotic support.

Speaker Bio: Janelle Clark is currently an assistant professor in the Mechanical Engineering Department at the University of Maryland, Baltimore County leading the Tactile and Robotic Assistance (TARA) Lab. She received her Ph.D. in Mechanical Engineering at Rice University, Houston, TX, USA in 2022 in the Mechatronics and Haptic Interfaces (MAHI) Lab and postdoctoral fellowship in the Human-Robot Interaction Laboratory in the Minor School of Computer Science at the University of Massachusetts Lowell, Lowell, MA, USA. Her research focuses on user-centric design and haptic interfaces for human-robot interactions in instances of shared embodiment, with application domains such as prosthetics, exoskeletons, teleoperation, and virtual reality. Her work combines principles from psychophysics, contact mechanics, robotics, hardware design, and controls to investigate physiological contributions to haptic perception and the creation of intuitive human-in-the-loop control systems.

Recording link: TBA

Abstract: https://uofi.box.com/s/uvzotk69o6kabmc9f8iyzirot7kcssyd
Mobility in robots is usually achieved by a few common means; with a few exceptions these are wheels or legs in ground robots, propellers in flying robots, articulated tails or fins and propellers in swimming robots or flapping flagella in micro swimmers. Nonlinear dynamics can provide insights into alternative means of generating efficient mobility. The talk will present several examples of locomotion produced by the interplay of variations in the inertia tensor, constraints (holonomic and nonholonomic) and periodic actuation. The actuation of such robots is achieved by means of internal actuators that do not directly interact with the environment. Periodic motion or spin of an internal body such as a rotor can transfer high frequency reaction forces and moments that in turn can produce oscillations of flexible structures like tails in a fish-like robot and in legs or cilia in a soft robot. Further these spin-generated forces modulate the forces at surfaces producing discontinuous phenomenon like slipping and jumping. In the low Reynolds number regime, spin actuation can produce propulsion of microswimmers and spinning swimmers can manipulate small particles in a contactless manner. The talk will demonstrate this framework with a spin driven swimming robot which has a locomotion efficiency approaching that of several species of fish, a spin driven pipe crawling robot, a spin driven jumping robot and a spin driven microswimmer and particle manipulator. Spin is all one needs.

Speaker Bio:
Phanindra Tallapragada is an associate professor of mechanical engineering at Clemson University. He obtained his Ph.D in Engineering Mechanics from Virginia Tech in 2010 and did post doctoral research at the University of North Carolina Charlotte. Earlier he obtained his B.Tech and M.Tech in Civil Engineering from the Indian Institute of Technology, Kharagpur. He joined Clemson University as an assistant professor in 2013. His research interests are in dynamical systems and bioinspired locomotion related to terrestrial motion, fish-like swimming, low Reynolds number swimming and operator methods for transport and manipulation in dynamical systems.

Recording link: https://uofi.box.com/s/2vhdeahk5sz2jn2l69hsvlatm3r5sh3a

Abstract:
We are witnessing a fascinating era in the evolution of autonomous robots. The integration of camera vision, AI-based perception, and advanced actuation and sensing technologies with off-the-shelf traditional robotic manipulators featuring multiple rigid degrees of freedom is nearing human-like precision. However, this precision, accuracy, and robustness are often compromised when the robot’s structure deviates from such traditional designs. Interestingly, around 90% of natural organisms are invertebrates, lacking a rigid backbone and exhibiting fluid, non-segmented morphologies. Creatures like snakes, octopus tentacles, and elephant trunks skillfully manipulate delicate objects without causing damage while still striking prey with remarkable precision. This raises a critical question: what fundamental advances in material structure design, actuation, sensing and perception, and controls are necessary to replicate these capabilities in bio-inspired robotic systems? This talk will showcase recent advancements in these areas for soft continuum manipulators, highlighting their potential applications in agriculture and healthcare. But are these advances sufficient to survive the axe, or will soft robots become obsolete before reaching the plateau of productivity?

Speaker Bio:
Girish Krishnan is an Associate Professor at the University of Illinois at Urbana-Champaign. He is also a Health Innovation Professor at the Carle Illinois College of Medicine. He obtained his PhD from University of Michigan at Ann Arbor and masters from the Indian Institute of Science. His research is in the intersection of design and robotics, and particularly in soft robotics with applications in agriculture and healthcare. Prof. Krishnan is the recipient of the 2015 NSF Early Career award (CAREER), 2016 UIUC council award for excellence in advising and 2017 Freudestein Young Investigator award (ASME). He has published around 40 peer-reviewed journals, 50 conference proceedings, and holds four patents.

Recording link: https://uofi.box.com/s/8eygnt0cs75vedh30jbd2jxf6634zx7q

Abstract:
Talk 1:
Teleoperation allows people to extend their perception and interaction capabilities to remote locations in cases where distance or safety constraints prevent them from physically traveling. While humans are extremely adept at a wide variety of manipulation tasks, transferring these skills through a robot has remained an open challenge for over half a century. Given constraints on an operator’s total cognitive load, high degree-of-freedom systems face a tradeoff between flexibility and precision. Giving the operator direct control over every joint on the robot provides them with the maximum flexibility to complete many kinds of tasks, but the high cognitive load associated with such an interface makes it impossible to properly coordinate the joints for tasks requiring high precision. In contrast, shared control systems that let the operator control the robot at a higher level of abstraction reduce this cognitive load and can improve the operator’s precision, but simultaneously reduce the flexibility of the interface to accomplish tasks for which it was not explicitly designed. Towards the goal of creating truly telepresent systems, I will discuss several learning and optimization techniques to customize robotic teleoperation interfaces to specific users and tasks, allowing operators to more naturally express their intent and enabling greater transfer of their manipulation skills.
Talk 2:
Humanoid robots have the potential to perform physically demanding tasks in industries such as manufacturing, construction, and healthcare. However, achieving precise and dynamic mobile manipulation—coordinating whole-body motion to interact forcefully with the environment—remains a challenge. Teleoperation provides a promising opportunity to bridge this gap by embedding human planning, coordination, and sensorimotor skills into robot control loops while also enabling rich data collection to enhance learning-based control policies. Yet, traditional teleoperation interfaces lack intuitive hands-free control and fail to provide adequate feedback, making it difficult for operators to simultaneously manage contact-rich locomotion and manipulation in real-time. In this talk, I will discuss the development of a bilateral teleoperation framework for controlling a wheeled humanoid robot for dynamic mobile manipulation using motion retargeting strategies, whole-body control, and a human-machine interface. The system allows operators to seamlessly switch between position and force control modes, balancing precision and compliance across different tasks. Additionally, haptic feedback improves operator awareness by synchronizing human-robot motion and conveying contact forces and moments from the robot’s interactions with the environment.

Speaker Bio:
Speaker 1:
Patrick Naughton is a PhD student at UIUC advised by Professor Kris Hauser. His work focuses on designing fluent interfaces for teleoperated manipulation using a combination of machine learning and optimization-based approaches. He previously received his Bachelor’s degree in Electrical Engineering at Washington University in St. Louis.
Speaker 2:
Amarty Purushottam is a PhD student at UIUC advised by Professor Joao Ramos & Professor Katie Driggs-Campbell. His work focuses on bilateral teleoperation, hybrid motion-force control, and developing humanoid robots. He previously received his Bachelor’s and Master’s degrees in Electrical Engineering at UIUC.

Recording link: https://uofi.box.com/s/4hj1sk4oyqfuo107atasai79xd0oyduv

Abstract:
Reliable interactions among robots require safety and resilient operational guarantees that adapt to real-world conditions. However, most current algorithm designs are rigidly optimized for well-defined and predictable environments. This not only renders pre-computed guarantees brittle under uncertainties and failures, but also incentivizes overly conservative robot behaviors that compromise primary task performance. In this talk, I will present our recent work harnessing safety and resilience for robust and jointly optimized multi-agent autonomy. I will first discuss a control-theoretic method to enforce minimally disruptive safe robotic behaviors under uncertainties, and how to leverage the expressivity of control-theoretic analysis to inspire risk-aware, decentralized multi-agent coordination in autonomous driving. In the presence of incomplete information, I will then show a sample-efficient reinforcement learning framework that allows a robot to safely explore with certified predictive confidence while simultaneously optimizing task performance. Finally, I will turn to resilient autonomy through the example of communication-aware multi-robot coordination. A set of communication-motion co-design frameworks will be demonstrated that allow robots to (i) optimize low-level motion constraints for system-level property guarantees, (ii) utilize data for compositional constraints learning in a realistic environment, and (iii) endure or recover from unexpected robot failures, leading to co-optimized system resilience and task efficiency.

Speaker Bio:
Wenhao Luo is an assistant professor in computer science at the University of Illinois Chicago. His research lies at the intersection of robotics, control theory, artificial intelligence, and machine learning. Specifically, his group develops principled methods for robust and interactive autonomy that enable robots to safely and effectively collaborate with each other and with humans in the physical world. Wenhao received his M.S. and Ph.D. in Robotics from Carnegie Mellon University in 2016 and 2021, respectively, and a B.E. degree with honors from Central South University in 2012. Before joining UIC, he was an assistant professor at UNC Charlotte (2021-2024) and a research intern at Microsoft Research in the summer of 2019 and 2020. Wenhao’s research has received best paper awards and nominations at multiple conferences, including AAMAS, IFAC CPHS, and workshops at ICRA, IROS, and IJCAI. He was also selected for AAAI New Faculty Highlights at AAAI 2023 and RSS Pioneers at RSS 2021. His work has been supported by NSF, USDA, DARPA, ONR, and AFOSR.

In-person event in CSL Studio 1232, no recording

Abstract:
The Multi-Scale Robotics & Automation Lab (MSRAL) at Purdue University performs cutting-edge research on robotic and automation systems at various length scales: macro-scale (cm to m), meso-scale (~100’s of um to a few mm’s), and micro-scale (10’s of um to 100’s of um). All the developed systems are designed to interact with the environment in unique ways. In this talk, I will discuss some recent MSRAL microrobotics projects on different types of families of wireless mobile microrobots driven by external magnetic fields that we have developed over the years and some recent work in agricultural robotics and mobile manipulation.

Speaker Bio:
David J. Cappelleri is the Assistant Vice President for Research Innovation in the Office of Research, B.F.S. Schaefer Scholar & Professor in the School of Mechanical Engineering, and Professor in the Weldon School of Biomedical Engineering (by courtesy) at Purdue University. Prof. Cappelleri founded the Multi-Scale Robotics & Automation Lab (MSRAL) that performs cutting-edge research on robotic and automation systems at various length scales. His research interests include mobile microrobotics for biomedical and manufacturing applications, surgical robotics, automated manipulation and assembly, and unmanned aerial and ground robot design for agricultural applications. Prof. Cappelleri is the Purdue site director for the NSF Engineering Research Center on the Internet of Things for Precision Agriculture (IoT4Ag). Prof. Cappelleri has received various awards, such as the NSF CAREER Award, Harvey N. Davis Distinguished Assistant Professor Teaching Award, the Association for Lab Automation Young Scientist Award, and is Fellow of the American Society of Mechanical Engineers (ASME). He received a Batchelor of Mechanical Engineering degree from Villanova University, a MS in Mechanical Engineering degree from The Pennsylvania State University, and a PhD degree in Mechanical Engineering & Applied Mechanics from the University of Pennsylvania.

Recording link: https://uofi.box.com/s/r3rher5wbcej7kxjkzxpo8vfggu8a328

Abstract:
Talk 1:
What if robotists had our own GPT-3 moment? Current Vision-Language-Action (VLA) models for robots have predominantly followed a zero-shot paradigm akin to GPT-2, achieving promising yet limited generalization without task-specific fine-tuning. In this talk, I propose a new hypothesis: by embracing an autoregressive framework that treats robot actions as next-token predictions, we can unlock powerful in-context imitation learning capabilities similar to GPT-3. Inspired by breakthroughs in large language models, this approach leverages demonstration trajectories as prompts, allowing robots to quickly adapt and perform novel tasks in diverse environments without additional training. This talk will focus on ICRT (https://icrt.dev/) and discuss how GPT-style autoregression could help make embodied intelligence more flexible and generalizable.
Talk 2:
Recent advances in imitation learning have significantly improved robots’ ability to acquire diverse manipulation skills. However, the popular imitation learning pipelines typically require extensive teleoperation data and often struggle to generalize across different visual and spatial conditions. In this talk, I will present two of our recent approaches to teaching robots new manipulation skills using only single-video demonstrations. We frame this as ‘open-world imitation from observation,’ where robots can learn and execute new manipulation skills by watching just one video of human demonstration, without requiring additional robot data collection. Our methods enable robots to successfully deploy the derived policies in previously unseen scenarios with diverse visual and spatial conditions.

Speaker Bio:
Speaker 1:
Letian (Max) Fu is a PhD student in Electrical Engineering and Computer Sciences at UC Berkeley, advised by Professor Ken Goldberg. He is also a research intern at NVIDIA, where he works with Jim Fan and Yuke Zhu. Max’s research focuses on building scalable multi-modal robot foundation models that generalize across tasks and environments. His work explores how to leverage the inductive biases of pre-trained vision and language models to accelerate robot learning.
Speaker 2:
Yifeng Zhu is a Ph.D Candidate in Computer Science at the University of Texas at Austin, co-advised by Prof. Peter Stone and Prof. Yuke Zhu. His research focuses on robot learning and robot manipulation. His thesis mainly focuses on how robots learning generalizable manipulation policies from a small amount of data. Yifeng was awarded RSS Pioneer in 2023.

Recording link: https://uofi.box.com/s/uooo78to7t89cr1wutbrznm93rt75ywk

Abstract:
Many of the recent advances in robot learning have been driven by scaling up computational, hardware and human resources without a fundamental improvement in the efficiency of learning. Consequently, robots remain prohibitively expensive to train and struggle at adapting to novel situations— a key hallmark of intelligence. In my talk, I will discuss how resource-rational learning enables robots to master complex tasks with minimal experience. I will present metareasoning algorithms that guarantee resource-rationality by deliberating about what to learn, and how to allocate resources like time, computation, and human assistance optimally. I will also show how simple algorithmic improvements, such as incorporating uncertainty awareness and state space factorization, can yield significantly more sample-efficient learning, ultimately enabling more scalable and adaptable robot behavior.

Speaker Bio:
Shivam Vats is a Postdoctoral Research Associate in the Department of Computer Science at Brown University, working with George Konidaris. His research focuses on developing efficient decision making algorithms that improve with minimal experience, enabling robots to continually learn on their own. His work spans robot manipulation, human-robot collaboration and multi-robot planning, with papers nominated for the Outstanding Interaction Paper Award at ICRA 2022 and Spotlight presentation at ICLR 2025. He earned his PhD in robotics from Carnegie Mellon University where he was co-advised by Maxim Likhachev and Oliver Kroemer.

Recording link: https://uofi.box.com/s/xpfes3as5p8iw9mu0q3721r335q8sue7

Abstract:
Safety filters are attractive as a modular framework for “robustifying” an autonomous system and have become increasingly popular in an era of black-box nominal control policies. However, constructing valid safety filters is a challenge for many real-world high-dimensional autonomous systems. Learning-based approaches to designing safety filters are one possible solution, though they suffer from learning errors and are particularly sensitive to distribution shifts in online deployment—an aspect that is not ideal for safety. In this talk, I will highlight two of the approaches we have taken to improve outcomes for neural safety filters:
For lower-dimensional systems (e.g. 4D-6D), we have introduced tools for refining neural safety filters. This takes advantage of the scalability of learning-based approaches with the safety guarantees of control-theoretic methods. Additionally, refinement allows for online adaptation in the face of novel information.
For higher-dimensional systems (7D-50D+) we show how semi-supervised learning using techniques from applied math and control theory can better guide the learning process. This work was recently nominated for best paper at the Learning for Dynamics and Control (L4DC) conference. Additionally, by parameterizing environmental conditions, we can show adaptability online.

Speaker Bio:
Sylvia Herbert is an Assistant Professor of Mechanical and Aerospace Engineering at the University of California, San Diego. Her research focus is to enable efficient and safe decision-making in robots and other complex autonomous systems while reasoning about uncertainty in real-world environments and human interactions. These techniques are backed by both theory and physical testing on robotic platforms.
Prior to joining UCSD, Professor Herbert received her PhD in Electrical Engineering from UC Berkeley, where she studied with Professor Claire Tomlin on safe and efficient control of autonomous systems. Before that, she earned her BS/MS at Drexel University in Mechanical Engineering. She is the recipient of the ONR Young Investigator Award, 2023 IROS Robocup Best Paper Award, 2025 Learning for Dynamics and Control (L4DC) Best Paper Nomination, Hellman Fellowship, UCSD JSOE Early Career Faculty Award, UC Berkeley Chancellor’s Fellowship, UC Berkeley Outstanding Graduate Student Instructor Award, and the Berkeley EECS Demetri Angelakos Memorial Achievement Award for Altruism.

Recording link: https://uofi.box.com/s/4c5b7saf7q7q0ceiev1hd7raf3qfz0nk

Abstract:
In this talk, I will outline the ingredients for enabling efficient robot learning. First, I will demonstrate how large vision-language models can enhance scene understanding and generalization, allowing robots to learn general rules from specific examples for handling everyday objects. Next, I will describe methods for leveraging equivariance to significantly reduce the amount of training data needed for learning from human demonstrations.
Moving beyond demonstrations, I will discuss how simulation can enable robots to learn autonomously. I will describe the challenges and opportunities of aligning differentiable simulators with reality, and also introduce methods for facilitating reinforcement learning with ‘black-box’ simulators. To further expand robot capabilities, we can adapt the hardware. In particular, I will demonstrate how differentiable simulation can be used for learning tool morphology to automatically adapt tools for robots. I will also outline experiments with new affordable and robust sensors. Finally, I will share plans for our new project on co-design of hardware and policy learning, which will leverage global optimization, rapid prototyping, and real-to-sim to jointly search the vast space of hardware designs and reinforcement learning methods.

Speaker Bio:
Rika Antonova is an Associate Professor at the University of Cambridge. Her research interests include data-efficient reinforcement learning algorithms, robotics, active learning & exploration​. Earlier, Rika was a postdoctoral scholar at Stanford University upon receiving the Computing Innovation Fellowship from the US National Science Foundation. Rika completed her PhD at KTH, and earlier she obtained a research Master’s degree from the Robotics Institute at Carnegie Mellon University. Before that, Rika was a senior software engineer at Google.

Recording link: https://uofi.box.com/s/igtm12fbuhfpfe3s2ru25ngy8nj5smrf

Abstract:
To transform our lives, autonomous systems need to interact with other agents in complex shared environments. For example, autonomous cars need to interact with pedestrians, human-driven cars, and other autonomous cars. Autonomous delivery drones need to navigate in the aerial space shared by other drones, or mobile robots in a warehouse must navigate in the factory space shared by robots. The multi-agent nature of such application domains requires us to develop a systematic methodology for enabling efficient interactions of autonomous systems across various applications. In this talk, I will first focus on game-theoretic planning and control for robots. To reach intelligent robotic interactions, robots must account for the dependence of agents’ decisions upon one another. I will discuss how game-theoretic planning and control enables robots to be cognizant of their influence on other agents. I will present our recent results on leveraging the structure that is inherent in interactions to develop efficient motion planning algorithms which are suitable for real-time operation on robot hardware. In the second part of the talk, I will focus on how robots can learn and infer the intentions of their surrounding agents to account for agents’ preferences and objectives. Currently, robots can infer the objectives of isolated agents within the formalism of inverse reinforcement learning; however, in multi-agent domains, agents are not isolated, and the decisions of all agents are mutually coupled. I will discuss a mathematical theory and numerical algorithms for inferring these interrelated preferences from observations of agents’ interactions.

Speaker Bio:
Negar Mehr is an assistant professor in the Department of Mechanical Engineering at the University of California, Berkeley. Before that, she was an assistant professor of Aerospace Engineering at the University of Illinois Urbana-Champaign. She was a postdoctoral scholar at Stanford Aeronautics and Astronautics department from 2019 to 2020. She received her Ph.D. in Mechanical Engineering from UC Berkeley in 2019 and her B.Sc. in Mechanical Engineering from Sharif University of Technology, Tehran, Iran, in 2013. She is a recipient of the NSF CAREER Award. She was awarded the IEEE Intelligent Transportation Systems best Ph.D. dissertation award in 2020.

Recording link: No recording

Abstract:
Optimal control problems for dynamic legged locomotion are difficult to solve while respecting real-time constraints. Identifying a technical strategy for a general purpose humanoid product requires careful thinking about state-of-the-art and product needs. Fundamental to both is identifying the correct problem to solve given all the constraints at hand. This talk will explore these topics and provide a perspective on how researchers and industry professionals can learn from one another about how to solve hard problems.

Speaker Bio:
Joe Norby is the lead for the Motion Control and Planning Team and the acting Director of Controls at Apptronik. He received the B.S. degree in mechanical engineering from the University of Notre Dame in 2016, and the Ph.D. degree in mechanical engineering from Carnegie Mellon University in 2022. His graduate work focused on global and local motion planning for autonomous agility in legged robots. At Apptronik, he specializes in navigation, dynamic locomotion, contact-rich manipulation, and real time controls for humanoid robots that perform useful work

Recording link: https://uofi.box.com/s/xr3ppgywubvs34v8gg54ajs2abadyfgj

Abstract:
Can we optimize the coordinated motion of a fleet of robots, when even for two robots we know so little? Can we quickly decide if one object could cage another? Is there an effective way to explain why we failed to find an assembly plan for a new product design? We review these and other challenging problems at the intersection of robotics and computational geometry—let’s call this intersection Geobotics. What is common to most of these problems is that the prevalent algorithmic techniques used in robotics do not seem suitable for solving them, or at least do not suggest quality guarantees for the solution. Solving some of them, even partially, can shed light on less well-understood aspects of computation in robotics. Joint work with Mikkel Abrahamsen.

Speaker Bio:
Dan Halperin received his Ph.D. in Computer Science from Tel Aviv University, after which he spent three years at the Computer Science Robotics Laboratory at Stanford University. He then joined the Department of Computer Science at Tel Aviv University, where he is currently a full professor and for two years was the department chair. Halperin’s main field of research is Computational Geometry and Its Applications. Application areas he is interested in include robotics, automated manufacturing, algorithmic motion planning, and 3D printing. A major focus of Halperin’s work has been in research and development of robust geometric software, in collaboration with a group of European universities and research institutes: the CGAL project and library, which recently won the SoCG test of time award. Halperin was the program-committee chair/co-chair of several conferences in computational geometry, algorithms and robotics, including SoCG, WAFR, ESA, and ALENEX. Halperin is an ACM Fellow and an IEEE Fellow.

Recording link: No recording

Abstract:
Manipulation is an essential skill enabling robots to physically engage them in real-world tasks. Being a topic involving multiple sub-problems, including planning, control, learning, design, and estimation, robot manipulation still remains a highly challenging field after decades of extensive research. Among others, one major issue always associated with manipulation systems is that it is hard to transfer most of the “successful” solutions from labs to real-world applications, either they be learning-based or analytical approaches, due to physical and perception uncertainties etc. In other words, we have been making too many wishful assumptions in the development of robot systems, such as that we will have enough data for training, we will have good sensors, etc. In this talk, I will focus on the exploration of common wishful assumptions, their impacts, as well as how we can develop robots with less of such assumptions. Research topics including robot self-identification, interactive perception, manipulation funnels, caging-based manipulation, and end-effector design will be discussed.

Speaker Bio:
Kaiyu Hang is an Assistant Professor of Computer Science at Rice University, where he leads the Robotics and Physical Interactions Lab. He is broadly interested in robotic systems that can physically interact with other robots, people, and the world. By developing algorithms in optimization, planning, learning, estimation, and control, his research is focused on efficient, robust, and generalizable manipulation systems, addressing problems that range from small scale grasping and in-hand manipulation, to large scale dual-arm manipulation, mobile manipulation, and multi-robot manipulation.

Recording link: https://uofi.box.com/s/8milojak13qu58qtyevtn9unjmb8dpjy

Abstract:
The rapid commoditization of high performance hardware for compute, sensing and actuation, paired with recent incredible breakthroughs in data driven methods for robot control, make a wave of new robot applications seem imminent. Yet, the long-standing vision of widespread real-world use of cutting edge robotics in areas where they are desperately needed, is still far from being fulfilled. Why is that, and how can we bridge the gap from what seems possible in the lab today to deployments in actual use cases that can have a positive impact on society? This talk describes TRI’s approach to answering this question with a multipath approach that spans various technologies and timespans. It also highlights several of TRI’s robotics research achievements and recent progress in autonomous mobile manipulation.

Speaker Bio:
Dr. Lukas Kaul is a Senior Research Scientist with the robotics division at the Toyota Research Institute in Los Altos, California. His current work revolves around hardware system design and algorithms for autonomous mobile manipulators, with a focus on the challenges of real-world deployability. He received his PhD in computer science from the Karlsruhe Institute of Technology (KIT) in 2019 for a thesis on efficient methods for humanoid balancing. Prior to that, he received his B.Sc. and M.Sc. in mechanical engineering from KIT, where his work spanned topics ranging from control circuits for high-voltage micro vibratory conveyors to large-scale mapping of underground caves with drones. Passionate about making robotics ever more accessible, he published his book “Practical Arduino Robotics”, a guide to getting started with building robots, in 2023.

Recording link: https://uofi.box.com/s/2l2m8pmvij78d8j6qrm9hbpwvazopzrd

Recording link: https://uofi.box.com/s/tr1l89du3fzx55xji281pj55fpaaqpcg

Abstract:
Contraction theory provides an analytical tool for studying differential dynamics of nonlinear systems under a contraction metric, the existence of which results in a necessary and sufficient characterization of the incremental exponential stability of multiple solution trajectories with respect to each other. More intuitively, it defines a systematic way to measure the distance of the systems’ current performance to their ideal, exponentially decreasing in time. It is increasingly recognized that the concept of contraction is fundamental in driving systems to their ideal counterparts both in physics-informed and data-driven systems. This talk gives a brief mathematical overview of contraction theory, with some recent efforts in generalizing the ideas to broader problem settings. Its practical impact is demonstrated through applications in several joint projects with NASA-JPL.

Speaker Bio:
Hiroyasu Tsukamoto is an Assistant Professor of Aerospace Engineering at the University of Illinois at Urbana-Champaign and the director of the ACXIS Laboratory (Autonomous Control, Exploration, Intelligence, and Systems). Prior to joining Illinois, he was a Postdoctoral Research Affiliate in Robotics at the NASA Jet Propulsion Laboratory, where he contributed to the Science-Infused Spacecraft Autonomy for Interstellar Object Exploration and Multi-Spacecraft Autonomy Technology Demonstration projects. He received his Ph.D. and M.S. in Space Engineering (Autonomous Robotics and Control) from Caltech in 2018 and 2023, respectively, and his B.S. degree in Aeronautics and Astronautics from Kyoto University, Japan, in 2017. He is the recipient of several awards, including the William F. Ballhaus Prize for the Best Doctoral Dissertation in Space Engineering at Caltech and the Innovators Under 35 Japan Award from MIT Technology Review. More info: https://hirotsukamoto.com

Recording link: https://uofi.box.com/s/pcadg90t8c63qbq8h2oyphyt234l5tra
Slides llink: https://uofi.box.com/s/pcadg90t8c63qbq8h2oyphyt234l5tra

Abstract:
Exoskeletons have been proposed to augment, assist, and rehabilitate motion. The efficacy of an exoskeleton in supporting the designed goals is affected by how a person moves with and uses the exoskeleton. This fluency between the human operator and exoskeleton is affected by the alignment between dimensions of the person and exoskeleton, but is also affected by the manner in which the exoskeleton is integrated in the perception-cognition-action decision process of the operator. In this talk, I will highlight studies that examine the interactions between sensory perception, executive function, and motor action selection while wearing a lower-body active exoskeleton during goal-oriented tasks.

Speaker Bio:
Leia Stirling is an Associate Professor in the Industrial and Operations Engineering Department and Robotics Department, an Affiliate Faculty in the Space Institute, and the University of Michigan Center for Occupational Health and Safety Engineering (COHSE) Director of Occupational Safety Engineering and Ergonomics. Her research group brings together methods from human factors, biomechanics, and robotics to understand the physical and cognitive interactions for goal-oriented human task performance and to support operational decision making that relies on manual task performance. These goals may include reducing musculoskeletal injury risks, supporting telehealth, and improving technology usability. Her research has been funded by the National Science Foundation, NASA, NIOSH, Boeing, and the U.S. Army.

Recording link: https://uofi.box.com/s/ptc2h7k9r27dl957dw2g3e8woeeenkrh

Abstract:
Soft robots made of compliant materials are adaptive to environment, robust to impact, and lightweight, making them safe for human-centered applications. Fluidic robots, as one of the most commonly used soft robots, have been widely applied in wearable technologies, including assistive devices, therapeutic tools, and human-machine interfaces. However, the power and control systems for fluidic robots generally are not only rigid, but also ten times heavier than the actuators, significantly limiting the robots’ mobility, flexibility, and safety to human. In this talk, I will introduce compliant, lightweight, and compact transducing systems that have enabled actuation and control of fluidic actuators. I will start by introducing a power-dense electrically-responsive soft actuator that is centimeter scale in size and weighs only 0.1g. Then I will present centi-meter scale soft valves and peristaltic pumps that employ these soft actuators as the core components. I will discuss the working mechanism and design criteria of these valves and pumps, and demonstrate their capabilities to generate and control fluidic flows. Lastly, I will demonstrate closed-loop control of a bending hydraulic actuator with the aforementioned soft valves and pumps. These soft actuators and transducing systems will serve as the foundation for my future work on untethered soft robotic systems for everyday applications, paving the way for advancements in wearable and implantable technologies.

Speaker Bio:
Siyi Xu is an Assistant Professor of Mechanical Science and Engineering University of Illinois at Urbana-Champaign and the director of the WISER Laboratory (Wearable and Implantable Soft Electronics and Robotics Lab). Prior to joining UIUC, she was a postdoctoral fellow at the Querrey Simpson Institute for Bioelectronics at Northwestern University, where she worked on implantable sensors for cardiac output flow and drug delivery monitoring for chemotherapy patients. She received her PhD in Mechanical Engineering at Harvard University in 2022. Her research focuses on developing soft robotic platforms equipped with compliant, lightweight, and compact sensing and transducing systems for human-centered applications. Siyi received her bachelor’s degree in Materials Science and Engineering from the University of Illinois at Urbana-Champaign. She was selected as a Rising Star in Mechanical Engineering in 2021. She is also a co-recipient of the IEEE Transactions on Robotics King-Sun Fu Memorial Best Paper Award.

Recording link: https://uofi.box.com/s/84g0o3mh73j9llcd5d77p2lv5yoj17nf

Recording link: https://uofi.box.com/s/rlw8msdrgosqtmzu433gtwn2a3yr31ji
Recording link: TBA

Abstract:
Commercially available bionic limbs have been far behind the state-of-the-art research that has been developed at academic institutions around the world. PSYONIC’s Ability Hand was developed to take advances in soft robotics and sensorimotor prostheses and make them accessible to humans and robots. The Ability Hand is an FDA-registered multiarticulated bionic hand that is the fastest on the market, robust to impacts, and gives users touch feedback. It is also covered by Medicare in the US. As a robotic manipulator, it is currently in use by NASA, Meta, Apptronik, Mercedes-Benz, and some of the top automakers, manufacturers, and brain-mother interface companies globally. This talk will detail the development of the Ability Hand, its current capabilities, and further advancements in bionic limbs that will be coming in the near future.

Speaker Bio:
Dr. Akhtar received his Ph.D. in Neuroscience and M.S. in Electrical & Computer Engineering from the University of Illinois at Urbana-Champaign in 2016. He received a B.S. in Biology in 2007 and M.S. in Computer Science in 2008 at Loyola University Chicago. His research is on motor control and sensory feedback for prosthetic limbs, and he has collaborations with the Center for Bionic Medicine at the Shirley Ryan AbilityLab, the John Rogers Research Group at Northwestern University, and the Range of Motion Project in Guatemala and Ecuador. In 2021, he was named as one of MIT Technology Review’s top 35 Innovators Under 35 and America’s Top 50 Disruptors in Newsweek.

Recording link: No recording

Abstract:
Many robotics problems have transition dynamics that are symmetric in SE(2) and SE(3) with respect to rotation, translation, scaling, reflection, and other transformations. In these situations, any optimal policy will also be symmetric over these transformations. In this talk, we leverage this insight to improve the sample efficiency of policy learning by encoding the symmetries directly into the neural network model using group invariant and equivariant layers. The result is that we can learn non-trivial visuomotor control policies with very little experience. In some cases, we can learn good policies from scratch by training directly on real robotic hardware in real time. We apply this idea both to imitation learning and reinforcement learning and achieve state of the art results in both cases.

Speaker Bio:
Rob Platt is an Associate Professor in the Khoury College of Computer Sciences at Northeastern University and a Faculty Fellow at BDAI. He is interested in developing robots that can perform complex manipulation tasks alongside humans in the uncertain everyday world. Much of his work is at the intersection of robotic policy learning, planning, and perception. Prior to coming to Northeastern and BDAI, he was a Research Scientist at MIT and a technical lead at NASA Johnson Space Center.

Recording link: https://uofi.box.com/s/rt3f9q3dohedfr99sad62p1ijsoog505

Abstract:
Kinematic singularities in robots are generally problematic.  There are two main types of singularities.  For the first type, a robot loses its ability to exert forces in a certain direction.  For the second type, a robot loses its ability to move in a certain direction.  Therefore it is best that singularities be avoided.  This presentation takes a backward approach.  Singularities are embraced, and we explore what extra functionalities might be obtained by purposefully designing singularities into the configuration space.  This act of design is nonelementary.   We take a computational approach, one based in root-finding.  The root-finding problems posed can be huge.  Therefore, we have pushed the bounds in coming up with more capable root finding algorithms.  Results are packaged into visualizations suitable for design space exploration.

Speaker Bio:
Mark Plecnik is an assistant professor in the Department of Aerospace and Mechanical Engineering at the University of Notre Dame. His research is grounded in the computational design of robot geometries. His current focus is in forming the prevailing dynamics of a robot by shaping its configuration space rather than depending purely on motor control to achieve desired motions. Plecnik received the NSF CAREER Award in 2022. He is a senior member of IEEE.

This event will consist of short faculty and student presentations to share their research insights, registration link: https://docs.google.com/forms/d/e/1FAIpQLSdNLdfT4BjZgmL5jl2jUjIB6MBriWhgCrHavkk25HLn9m1Lgg/viewform.

Recording link: https://uofi.box.com/s/ib5f5v9dxhv4rgg7inpjynkqs6tpttv4

Abstract:
Human-robot collaboration has the potential to improve people’s lives by enabling robots to provide timely assistance. For collaborative robots to live up to their potential, they must be able to learn from people and adapt to their human partners’ needs. While learning from demonstration has been successful for learning robot skills, demonstrations are only one way that people teach each other. In this work, we present an extended learning framework that formalizes four teaching archetypes. We empirically show that the type of feedback people provide has different benefits to the learner and costs to the teacher. We also describe how robots can actively learn by selecting the best feedback type at any given moment. Finally, we discuss how robots should communicate their own models to help people teach more effectively. 

Speaker Bio:
Dr. Henny Admoni is an Associate Professor in the Robotics Institute at Carnegie Mellon University, where she leads the Human And Robot Partners (HARP) Lab. Dr. Admoni’s research interests include human-robot interaction, assistive robotics, and nonverbal communication. Dr. Admoni holds a PhD in Computer Science from Yale University, and a BA/MA joint degree in Computer Science from Wesleyan University.

Recording link: https://uofi.box.com/s/7zsydpq40v2nx5wlc1u9ugmzgj1trn44

Abstract:
For all the promise of big-data machine learning, what will happen when robots deploy to our homes and workplaces and inevitably encounter new objects, new tasks, and new environments? If a solution to every problem cannot be pre-trained, then robots will need to adapt to this novelty. Can a robot, instead, spend a few seconds to a few minutes gathering information and then accomplish a complex task? Why does it seem that so much data is required, anyway? I will first argue that the hybrid or contact-driven aspects of manipulation clashes with the inductive biases inherent in standard learning methods, driving this current need for large data. I will then show how contact-inspired implicit learning, embedding convex optimization, can reshape the loss landscape and enable more accurate training, better generalization, and ultimately data efficiency. Finally, I will present our latest results on how these learned models can be deployed via real-time multi-contact MPC for dexterous robotic manipulation, where the robot must autonomously make and break contact and initiate stick-slip transitions.

Speaker Bio:
Michael Posa is an Assistant Professor in Mechanical Engineering and Applied Mechanics at the University of Pennsylvania. He leads the Dynamic Autonomy and Intelligent Robotics (DAIR) lab, a group within the Penn GRASP laboratory. His group focuses on developing computationally tractable algorithms to enable robots to operate both dynamically and safely as they interact with their environments. Michael received his Ph.D. in Electrical Engineering and Computer Science from MIT in 2017 and received his B.S. in Mechanical Engineering from Stanford University in 2007. Before his doctoral studies, he worked as an engineer at Vecna Robotics. He received the NSF CAREER Award in 2023, the RSS Early Career Spotlight in 2023, a Google Faculty Research Award, and a Young Faculty Researcher Award from the Toyota Research Institute. His work has also received awards recognition at TRO, ICRA, Humanoids, and HSCC.

Recording link: https://uofi.app.box.com/s/9rdz2y3wdoheet9e214r5wnd0a05o4m6

Abstract: TBA

Speaker Bio: TBA

Recording link: https://uofi.box.com/s/g600o7q3d2mmcmjsojgytsutnhm7y4sa

Abstract:
Legged robots have the potential to assist humans with a wide range of real-world tasks in dynamic, unstructured environments. While legged robot operation in inertial environments has been extensively studied, legged locomotion in non-inertial settings (e.g., ships and oil platforms) remains a new robot functionality that has not been solved. This new functionality will empower legged robots to perform critical, high-risk tasks such as shipboard maintenance, inspection, firefighting, and fire suppression as well as surveillance and disinfection on moving public transportation vehicles. Yet, enabling reliable locomotion in a non-inertial environment presents substantial fundamental challenges in robot control due to the high complexity of the hybrid, time-varying physical interaction between the robot and the environment. Dr. Gu will present the current progress from her research group in creating new approaches of robot modeling, state estimation, and control that achieve provably robust quadrupedal and humanoid locomotion in non-inertial environments. She will also report in-lab and shipboard test results that reveal major performance gaps of commercial legged robot systems in addressing dynamic shipboard environments

Speaker Bio:
Yan Gu received the B.S. degree from Zhejiang University, China, in 2011 and the Ph.D. degree from Purdue University in 2017, both in Mechanical Engineering. She joined the School of Mechanical Engineering at Purdue University as an Associate Professor in Fall 2022. Prior to joining Purdue, she was an Assistant Professor in the Department of Mechanical Engineering at the University of Massachusetts Lowell. Her research focuses on modeling, state estimation, planning, and control of legged locomotion and manipulation in highly dynamic and complex environments including non-inertial settings. Her research draws on nonlinear control theory, theory of hybrid systems, dynamics, and optimization to advance robot modeling, state estimation, and control. Dr. Gu is an Associate Editor for the IEEE/ASME Transactions on Mechatronics and IEEE Robotics and Automation Letters, as well as a Guest Editor for the IEEE Transactions on Robotics. She received the Young Investigator Program Award from the Office of Naval Research in 2023, the Faculty Early Career Development Program (CAREER) Award from the National Science Foundation in 2021, and Verizon’s 5G Robotics Challenge Award in 2019. Her research has been reported by various media such as Boston Globe, CNET, Robotics Business Review, and NPR’S WBUR.

Recording link: https://uofi.box.com/s/yn0r8r61lnslt3fvqq6qvqqiwmpyl9xv

Abstract:
Deploying a fleet of autonomous vehicles effectively is a challenging problem. A team of robots working together has the potential to outperform a large monolithic system because the team offers intrinsic resilience to individual failures and can inhabit large geographic areas while lowering the costs of individual robots. The team can also expand its capabilities by using heterogeneous robots with different abilities (e.g., sensors, effectors, dynamics, or computation). However, in order to bring these advantages to fruition, a user needs the ability to specify desired collective behaviors for the swarm while maintaining mechanisms for self-monitoring, self-reconfiguration, and self-repair. Using existing technologies, the job of tasking and controlling large numbers of systems often overwhelms human operators, as the number of variables in their decision-making process grows exponentially. Existing automated approaches for coordinating heterogeneous teams of robots either consider small numbers of agents, are application-specific, or do not adequately address common real-world requirements, e.g., strict deadlines or inter-task dependencies.
This talk will address these challenges by introducing scalable and robust algorithms for task-based coordination from high-level specifications to coordinate heterogenous teams. I will talk about a formal specification language, capability temporal logic (CaTL), to describe rich, temporal properties involving tasks requiring the participation of multiple agents with multiple capabilities, e.g., sensors or end effectors. Arbitrary missions and team dynamics can be jointly encoded as constraints in a mixed integer linear program, and solved efficiently using commercial off-the-shelf solvers. I will then walk through several hardware demonstrations and some computational results from deploying a planning and control system based on CaTL.

Speaker Bio:
Zachary Serlin is currently a technical staff research scientist in the Air, Missile, and Maritime Defense Technology Division at MIT Lincoln Laboratory. His research focuses on formal guarantees for artificial intelligence-based multi-agent autonomous systems. He has worked on large-scale logistics and command and control systems, automated strategy generation systems for heterogenous multi-agent coordination, and verifiable and resilient uncrewed platform planning and control systems. His work in coordinated heterogenous autonomy has been a finalist at the 2022 R&D 100 awards. He holds a BS and MS in mechanical engineering from Tufts University and a Ph.D. in mechanical engineering from Boston University.

Recording link: https://drive.google.com/drive/folders/1plW17-VtiBq75SIMcXztdiNzfmaJtP1A?usp=sharing

Abstract:
In the era of data-driven foundation models, dexterous manipulation sees the chance of being revolutionized by learning from data. However, as with all robotics tasks, we are facing a barrier: there is not much robotics data on hand, and there is not a clear way how we should get it. Efforts to collect very large datasets have required tremendous amounts of time and money; the great variety of robot embodiments and tasks also post a great challenge.
Recently, techniques like Imitation Learning and Reinforcement Learning have produced promising results in a number of dexterous manipulation tasks. In this seminar, we will discuss and debate two primary methods for acquiring dexterous manipulation data in IL and RL: teleoperation and simulation. Teleoperation offers environment realism and human expertise but relies heavily on human involvement. Simulation provides a safe, cost-effective environment for training but faces challenges in accurately representing the physical world. The seminar aims to delve into the strengths and limitations of each approach and discuss what are some promising future research directions.

Recording link: TBA

Abstract:
Commercially available bionic limbs have been far behind the state-of-the-art research that has been developed at academic institutions around the world. PSYONIC’s Ability Hand was developed to take advances in soft robotics and sensorimotor prostheses and make them accessible to humans and robots. The Ability Hand is an FDA-registered multiarticulated bionic hand that is the fastest on the market, robust to impacts, and gives users touch feedback. It is also covered by Medicare in the US. As a robotic manipulator, it is currently in use by NASA, Meta, Apptronik, Mercedes-Benz, and some of the top automakers, manufacturers, and brain-mother interface companies globally. This talk will detail the development of the Ability Hand, its current capabilities, and further advancements in bionic limbs that will be coming in the near future.

Speaker Bio:
Dr. Akhtar received his Ph.D. in Neuroscience and M.S. in Electrical & Computer Engineering from the University of Illinois at Urbana-Champaign in 2016. He received a B.S. in Biology in 2007 and M.S. in Computer Science in 2008 at Loyola University Chicago. His research is on motor control and sensory feedback for prosthetic limbs, and he has collaborations with the Center for Bionic Medicine at the Shirley Ryan AbilityLab, the John Rogers Research Group at Northwestern University, and the Range of Motion Project in Guatemala and Ecuador. In 2021, he was named as one of MIT Technology Review’s top 35 Innovators Under 35 and America’s Top 50 Disruptors in Newsweek.

Recording link: https://uofi.box.com/s/wtov1k9fb3x7p5zicy10wy8d25cu82bu

Abstract:
This talk introduces dynamic game theory as a natural modeling tool for multi-agent interactions ranging from large, abstract systems such as ride-hailing networks to more concrete, physically-embodied robotic settings such as collision-avoidance in traffic. We present the key theoretical underpinnings of dynamic game models for these varied situations and draw attention to the subtleties of information structure, i.e., what information is implicitly made available to each agent in a game. Thus equipped, the talk presents a state-of-the-art technique for solving these games, as well as a set of “dual” techniques for the inverse problem of identifying players’ objectives based on observations of strategic behavior.

Speaker Bio:
David Fridovich-Keil is an assistant professor at the University of Texas at Austin. David’s research spans optimal control, dynamic game theory, learning for control, and robot safety. While he has also worked on problems in distributed control, reinforcement learning, and active search, he is currently investigating the role of dynamic game theory in multi-agent interactive settings such as traffic. David’s work also focuses on the interplay between machine learning and classical ideas from robust, adaptive, and geometric control theory.

Recording link: TBA

Abstract:
Modern robots are highly capable in predictable environments, but often struggle in unstructured environments like granular media. However, loose terrains and granular environments are ubiquitous in our world, and making robots agile in these environments would open doors to applications in marine and space exploration, geotechnics, disaster response, and agriculture. In this presentation, I introduce our mole crab-inspired robot EMBUR (EMerita BUrrowing Robot), a legged robot capable of self-burrowing in granular media. I discuss current limitations for modeling such systems, and present my prior work on extending Granular Resistive Force Theory (RFT) to three dimensions. Lastly, I present demonstrations of tethered robotic teams in the field, and show how leveraging contact between the tether and natural objects can enable forceful maneuvers, even in complex terrains. I conclude my talk with a discussion of how the modeling techniques introduced can be extended to better inform robot design and control.

Abstract:
Laura Treers is a Postdoctoral Researcher in the Schools of Physics and Biological Sciences at Georgia Tech, working with Profs. Dan Goldman and Mike Goodisman. She received her PhD in Mechanical Engineering from UC Berkeley in summer 2023, working with Professor Hannah Stuart. She completed her undergraduate at MIT in 2018, also majoring in Mechanical Engineering. Her doctoral work focused on improving robotic capability in complex environments such as granular media, through novel mechanisms and modeling techniques. Her postdoctoral work is centered on manipulation and construction using cohesive granular materials, in both biological collectives (fire ants) and robots. She mentored undergraduate, masters, and junior PhD students throughout her time at Berkeley, is an active mentor for youth FIRST robotics programs, and feels strongly about helping students from all backgrounds and walks of life feel like they belong in STEM fields. In fall of 2024 she will start as an Assistant Professor of Mechanical Engineering at the University of Vermont (UVM).

Recording link: https://uofi.app.box.com/s/bdxeq2jpphz4xg43g77bz1jlwp526ah2

Abstract:
Recent advances in soft robotics motivate the design of multifunctional composites with actuation and perception capabilities. These functionalities are required for addressing long-standing challenges in soft robot control and achieving more sophisticated bioinspired behaviors. However, continued progress towards this vision is stymied by limitations in current materials and manufacturing methods. With these challenges in mind, I will present approaches for designing and fabricating soft robots from architected materials with distributed sensing capabilities. First, I will introduce 3D printed architectures called handed shearing auxetics for use as motorized soft actuators. I will then introduce approaches to sensorizing these actuators through fluidic innervation, where networks of empty, air-filled cavities provide distributed sensing through pressure measurements. Finally, I will discuss opportunities for using these systems for untethered locomotion and other embodiments of robotic materials.

Speaker Bio:
Ryan Truby is the June and Donald Brewer Junior Professor and an Assistant Professor of Materials Science and Engineering and Mechanical Engineering at Northwestern University. His research broadly aims to advance machine intelligence by material design. He and his team in the Robotic Matter Lab are currently developing novel soft actuators and sensors, rapid multimaterial 3D printing methods, and machine learning-based control strategies for soft and bioinspired robots. Ryan’s research also includes work in 3D printing vascularized tissue constructs, soft electronics, artificial muscles, and architected materials. Prior to Northwestern, Ryan was a Postdoctoral Associate at MIT’s Computer Science and Artificial Intelligence Lab, and he received his Ph.D. in Applied Physics from Harvard University. Ryan is the recipient of Young Investigator Program Awards from the Office of Naval Research and the Air Force Office of Scientific Research, the Outstanding Paper Award at the 2019 IEEE International Conference on Soft Robotics, and the Gold Award for Graduate Students from the Materials Research Society. His work at the materials-robotics interface has been supported by a Schmidt Science Fellowship and an NSF Graduate Research Fellowship.

Recording link: https://uofi.app.box.com/s/odlsgifpk52txkjje4oixq1kop2l3rcm

Abstract:
Humans have a strong intuitive understanding of the physical world. Through observations and interactions with the environment, we build a mental model that predicts how the world would change if we applied a specific action (i.e., intuitive physics). My research draws on insights from humans and develops model-based reinforcement learning (RL) agents that learn from their interactions and build predictive models that generalize widely across a range of objects made with different materials. The core idea behind my research is to introduce novel representations and integrate structural priors into the learning systems to model the dynamics at different levels of abstraction. I will discuss how such structures can make model-based planning algorithms more effective and help robots accomplish complicated manipulation tasks (e.g., manipulating an object pile, shaping deformable foam into a target configuration, and making a dumpling from the dough using various tools). Furthermore, I will demonstrate how structured scene representation facilitates the integration of large language models (LLMs), granting robots the ability to perform a large variety of everyday tasks specified in free-form natural language.

Speaker Bio:
Yunzhu Li is an Assistant Professor of Computer Science at the University of Illinois Urbana-Champaign (UIUC). Before joining UIUC, he collaborated with Fei-Fei Li and Jiajun Wu during his Postdoc at Stanford. Yunzhu earned his PhD from MIT under the guidance of Antonio Torralba and Russ Tedrake. His work stands at the intersection of robotics, computer vision, and machine learning, with the goal of helping robots perceive and interact with the physical world as dexterously and effectively as humans do. Yunzhu’s work has been recognized through the Best Systems Paper Award and the Finalist for Best Paper Award at the Conference on Robot Learning (CoRL). Yunzhu is also the recipient of the Adobe Research Fellowship and was selected as the First Place Recipient of the Ernst A. Guillemin Master’s Thesis Award in Artificial Intelligence and Decision Making at MIT. His research has been published in top journals and conferences, including Nature, NeurIPS, CVPR, and RSS, and featured by major media outlets, including CNN, BBC, The Wall Street Journal, Forbes, The Economist, and MIT Technology Review.

Recording link: https://uofi.app.box.com/s/31vwj904sbk7zthc62x3mi1imkrfwydi

Abstract:
My research philosophy revolves around elucidating the fundamental physics behind intriguing natural phenomena, leveraging cutting-edge technologies to engineer these principles, and applying them to develop innovative mechanisms that foster the safe, intelligent, and resilient coexistence of robotic systems with humans. This approach encompasses both the scientific and technological dimensions of robotics, emphasizing translational research. In this presentation, I will concentrate on our endeavors to design systems with adaptive morphology and embodied intelligence. These systems are adept at swiftly adapting to the ever-changing environmental conditions without imposing excessive computational demands on a central control unit. This also implies that such systems are capable of decentralizing certain calculations, entrusting them to the body itself. Additionally, I will introduce various topics within soft robotics, including soft underwater robots, soft-flying and locomotive robots, morphological designs for soft haptic interfaces, encompassing haptic sensing and haptic display, as well as safety measures and controls for drones and robot arms that rely on adaptive morphology and multimodal sensing.

Abstract:
Van Anh Ho (Senior Member, IEEE) received the Ph.D. degree in robotics from Ritsumeikan Unviersity, Kyoto, Japan, in 2012. Before that, he obtained the Bachelor degree in Electrical Engineering at Hanoi University of Science and Technology, Vietnam, in 2007; and Master degree in Mechanical Engineering in 2009 at Ritsumeikan University. He completed the JSPS Postdoctoral Fellowship in 2013 before joining Advanced Research Center Mitsubishi Electric Corp., Japan as a research scientist. From 2015 to 2017, he worked as Assistant Professor with Ryukoku University, Kyoto, Japan where he led a laboratory on soft haptics, soft modeling. From 2017, he joined the Japan Advanced Institute of Science and Technology (JAIST) for setting up a laboratory on soft robotics. His current research interests are soft robotics, soft haptic interaction, tactile sensing, grasping and manipulation, bio-inspired robots. He was recipient of the prestigious Japan Society for the Promotion of Science (JSPS) Research Fellowship for Young Scientist for his PhD course (DC) and postdoctoral fellowship. Ho was the recipient of 2019 IEEE Nagoya Chapter Young Researcher Award, and finalists of: Best System Paper at RSS 2023, Best Paper at IEEE SII (2016) and IEEE RoboSoft (2020). He is member of The Robotics Society of Japan (RSJ), and Senior Member of the IEEE. He is serving as Associate Editor for many international conferences, such as RSS, ICRA, IROS, RoboSoft; as well as for journals such as IEEE Transactions for Robotics (T-RO), IEEE Robotics and Automation Letters (RA-L), and Advanced Robotics. He is General Co-Chair of 2023 IEEE/SICE International Symposium on System Integration (SII), and General Chair of SII 2024.

Recording link: https://uofi.app.box.com/s/o0m0gn2i52h4l5hykpyb61zq8bp2gp1u

Abstract:
Guaranteeing effective exploration is a vital component in the success of remote robotic applications in ocean and space exploration, environmental monitoring, and search and rescue tasks. This talk presents a novel formulation of exploration that permits optimality criteria and performance guarantees for robotic exploration tasks. We define the problem of exploration as a coverage problem on continuous (infinite-dimensional) spaces based on ergodic theory and derive control methods that satisfy optimality and guarantees such as asymptotic coverage, set-invariance, time-optimality, and reachability in exploration tasks. Last, we demonstrate successful execution on a range of robotic systems and provide an outlook on applications to robot learning problems.

Abstract:
“Ian Abraham is an Assistant Professor in Mechanical Engineering with courtesy appointment in the Computer Science Department at Yale University. His research group is focused on developing real-time optimal control methods for data-efficient robotic learning and exploration. Before joining Yale, he was a postdoctoral researcher at the Robotics Institute at Carnegie Mellon University in the Biorobotics Lab. He received his PhD. and M.S. degrees in Mechanical Engineering from Northwestern University and the B.S. degree in Mechanical and Aerospace Engineering from Rutgers University. During his Ph.D. he also worked at the NVIDIA Seattle Robotics Lab where he worked on robust model-based control for large parameter uncertainty. His research interest lies at the intersection of robotics, optimal control, and machine learning with a focus on developing real-time embedded software for exploration and learning. He is the recipient of the 2023 Best Paper Award at the Robotics: Science and Systems conference, the 2019 King-Sun Fu IEEE Transactions on Robotics Best Paper award, the Northwestern Belytschko Outstanding Research award for his dissertation, and the 2023 NSF CAREER award.”

Recording link: https://uofi.app.box.com/s/7gvov5udxtbqeiqpbdsirhlx7cq5qo2u

Abstract:
Powered prostheses are capable of improving quality of life for 600,000 people with lower-limb amputations by enabling mobility with less human energy than traditional prostheses. However, current prosthesis control methods require many hours of heuristic tuning for every user and for every behavior. To develop a control method that applies across users, we construct model-based prosthesis control methods that rely solely on local information by translating formal nonlinear bipedal walking control methods to prostheses through a separable subsystem framework. On hardware, we realized the first model-dependent prosthesis controller that utilizes real-time force sensing to complete the model, improving tracking performance across terrains and subjects without additional tuning. Going forward, we will examine how assistive devices impact human biomechanical performance to guide the design of lower-limb assistive devices.

Speaker Bio:
Rachel Gehlhar Humann will be joining the UMN ME faculty in Spring 2024 as an assistant professor. Currently she is conducting her postdoctoral research in the Anatomics Lab at UCLA where she is studying how ankle stiffness impacts human biomechanics. She recently completed her PhD in mechanical engineering at the California Institute of Technology where she investigated nonlinear control methods for powered prostheses. Before Caltech, she graduated with a B.S. in mechanical engineering from the University of St. Thomas in St. Paul, MN.

Recording link: https://uofi.app.box.com/s/4y1vub1wauzznjw2ue5dslgeky2ombmf

Abstract:
Legged robots possess the unique advantage of negotiating unstructured terrains and cluttered environments by using their extremities to make discrete contact. The many real-world applications of the legged robot include disaster response, smart factory, home service and elderly care. My research endeavors to utilize model-based optimization methods to instill legged robots with their animal counterparts’ agility. This presentation provides an overview of the development of two custom robot platforms with highly dynamic capabilities. The first part focus on the formulation of a model predictive control (MPC) framework, which enables a quadruped robot to achieve dynamic maneuvers with a large angular excursion without singularity issues associated with Euler angles. The second part highlights contribution to the MIT Humanoid project, encompassing research thrusts in footstep adaptation, self-collision avoidance, and hardware experimentations.

Speaker Bio:
Yanran Ding is an assistant professor in the Robotics Department at the University of Michigan. His research goal is to realize agile motions on robot hardware that enables them to better provide physical services. Prior to joining Michigan, he worked as a postdoctoral associate at the MIT Biomimetic Robotics Lab. He earned his Ph.D. in the Mechanical Science and Engineering Department in the University of Illinois at Urbana-Champaign in 2021. He obtained his bachelor’s degree from Shanghai Jiao Tong University in 2015. He is the best student paper finalist at IROS 2017, and best paper finalist at TC on model-based optimization for robotics 2021.

Recording link: TBA

Abstract:
The significance of tactile sensing for enabling more intelligent interactions between robots and their environment is widely recognized; however, the rate of growth in the application of tactile sensing within the field of robotics has not met our expectations. The reasons vary. In this talk, I will talk about our effort in two directions: simulating tactile sensors and using an active action-perception loop to explore the environment through touch. I will introduce how we build the simulators and the algorithms, and discuss what we have learned from them so far. While most of the work I present will be about “what we see as a possible way to push forward” rather than providing an answer, I hope the talk will inspire more thoughts about robotic tactile perception and encourage more collaborations.

Speaker Bio:
Legged robots possess the unique advantage of negotiating unstructured terrains and cluttered environments by using their extremities to make discrete contact. The many real-world applications of the legged robot include disaster response, smart factory, home service and elderly care. My research endeavors to utilize model-based optimization methods to instill legged robots with their animal counterparts’ agility. This presentation provides an overview of the development of two custom robot platforms with highly dynamic capabilities. The first part focus on the formulation of a model predictive control (MPC) framework, which enables a quadruped robot to achieve dynamic maneuvers with a large angular excursion without singularity issues associated with Euler angles. The second part highlights contribution to the MIT Humanoid project, encompassing research thrusts in footstep adaptation, self-collision avoidance, and hardware experimentations.

Recording link: https://uofi.box.com/s/njuzotzy99a16jqrlwm73cs4p0golboo

Abstract:
To materialize the vision of intelligent robots in the real world, it is imperative that robots exhibit the ability to operate over extended durations and engage in collaborative endeavors across diverse scenarios, often alongside human counterparts within a shared environment. This presentation will elucidate our ongoing research within the field of lifelong collaborative autonomy, which serves as a pivotal step towards realizing these robotic capabilities. Firstly, I will present our work on context-aware self-reflective adaptation, which empowers lifelong robots to dynamically adapt not only to changes in their external surroundings but also to modifications in their own functionalities. Following this, I will delve into our research on adaptive peer-to-peer teaming within scenarios involving multi-robot formation control, tightly-coupled cooperation, and multimodal human-robot collaboration. Finally, the presentation will conclude with a discussion of potential open questions and unmet capabilities in lifelong collaborative robotics.

Speaker Bio:
Hao Zhang is an Associate Professor of Computer Sciences at the University of Massachusetts Amherst, where he leads the Human-Centered Robotics Laboratory. His research is centered around lifelong collaborative autonomy with a particular emphasis on robot adaptation, multirobot collaboration, and human-robot teaming. He is the recipient of an NSF CAREER Award, a DARPA Young Faculty Award (YFA), and a DARPA Director’s Fellowship. His research program benefits from the generous support by a diverse spectrum of sponsors, including NSF, DARPA, ARL, ONR, DOE, DOT, and industry partners. He consistently publishes in top robotics and AI conferences and journals, and has received multiple best paper awards and nominations.

Recording link: TBA

Abstract: TBA

Speaker Bio: TBA

Recording link: https://uofi.box.com/s/2gh8byltdvx5vyxihdqc9lie4kxnvv1q

Abstract:
New advances in robotics and autonomy offer a promise of revitalizing final assembly manufacturing, assisting in personalized at-home healthcare, and even scaling the power of earth-bound scientists for robotic space exploration. Yet, in real-world applications, autonomy is often run in the O-F-F mode because researchers fail to understand the human in human-in-the-loop systems. In this talk, I will share exciting research we are conducting at the nexus of human factors engineering and cognitive robotics to inform the design of human-robot interaction. In my talk, I will focus on our recent work on 1) enabling machines to learn skills from and model heterogeneous, suboptimal human decision-makers, 2) “white-box” that knowledge through explainable Artificial Intelligence (XAI) techniques, and 3) scale to coordinated control of stochastic human-robot teams. The goal of this research is to inform the design of autonomous teammates so that users want to turn – and benefit from turning – to the O-N mode.

Speaker Bio:
Dr. Matthew Gombolay is an Assistant Professor of Interactive Computing at the Georgia Institute of Technology. He was named the Anne and Alan Taetle Early-career Assistant Professor in 2018. He received a B.S. in Mechanical Engineering from the Johns Hopkins University in 2011, an S.M. in Aeronautics and Astronautics from MIT in 2013, and a Ph.D. in Autonomous Systems from MIT in 2017. Gombolay’s research interests span robotics, AI/ML, human-robot interaction, and operations research. Between defending his dissertation and joining the faculty at Georgia Tech, Dr. Gombolay served as technical staff at MIT Lincoln Laboratory, transitioning his research to the U.S. Navy and earning a R&D 100 Award. His publication record includes a best paper award from the American Institute for Aeronautics and Astronautics and the ACM/IEEE Conference on Human-Robot Interaction (HRI’22) as well as finalist awards for the best paper at the Conference on Robot Learning (CoRL’20) and best student paper at the American Controls Conference (ACC’20). Dr Gombolay was selected as a DARPA Riser in 2018, received the Early Career Award from the National Fire Control Symposium, and was awarded a NASA Early Career Fellowship.

Recording link: https://uofi.box.com/s/584lcogbawqfsysk5mraz9uks5w1nbh6

Abstract:
Legged robots show great promise in navigating environments that are difficult for conventional platforms, but they do not yet have the agility, endurance, and related physical ability to deliver on this potential. This talk presents an overview of three robots that address different aspects of legged mobility (organized in increasing numbers of legs). The monoped Salto-1P explores agile leaping to enable a small robot to clear large obstacles, simple bipeds investigate how walking scales to smaller sizes, and perception and control for quadruped Spirit-40 tackle walking through entanglements. Discussion of future directions (and potential avenues for collaboration) concludes the presentation.

Speaker Bio: Dr. Justin Yim is a new assistant professor at UIUC in the Mechanical Science and Engineering department. He received his Ph.D. in Electrical Engineering from the University of California, Berkeley and his B.S.E. and M.S.E. from the University of Pennsylvania. Prior to starting at UIUC, he was a Computing Research Association CIFellow 2020 postdoc with Aaron Johnson at Carnegie Mellon University. Justin Yim’s research interests are in the design and control of legged robots to improve performance and understand locomotion principles. For his dissertation work developing the jumping monopod robot Salto-1P, he received best paper and best student paper awards at the IEEE/RSJ IROS and IEEE ICRA conferences.

Recording link: https://uofi.box.com/s/jagw0mjnez37aiqfk00iqgueemw1ir7r

Abstract:
“Despite the incredible capabilities (speed, repeatability) of their hardware, most robot manipulators today are deliberately programmed to avoid dynamics – moving slow enough so they can adhere to quasi-static assumptions about the world. In contrast, people frequently (and subconsciously) make use of dynamic phenomena to manipulate everyday objects – from unfurling blankets to tossing trash, to improve efficiency and physical reach range. These abilities are made possible by an intuition of physics, a cornerstone of intelligence. How do we impart the same to robots? In this talk, I will discuss how we might enable robots to leverage dynamics for manipulation in unstructured environments. Modeling the complex dynamics of unseen objects from pixels is challenging. However, by tightly integrating perception and action, we show it is possible to relax the need for accurate dynamical models. Thereby allowing robots to (i) learn dynamic skills for complex objects, (ii) adapt to new scenarios using visual feedback, and (iii) use their dynamic interactions to improve their understanding of the world. By changing the way we think about dynamics – from avoiding it to embracing it – we can simplify a number of classically challenging problems, leading to new robot capabilities.”

Speaker Bio:
Shuran Song is an Assistant Professor in the Department of Computer Science at Columbia University. Before that, she received her Ph.D. in Computer Science at Princeton University, BEng. at HKUST. Her research interests lie at the intersection of computer vision and robotics. Song’s research has been recognized through several awards, including the Best Paper Awards at RSS’22 and T-RO’20, Best System Paper Awards at CoRL’21, RSS’19, and finalist at RSS, ICRA, CVPR, and IROS. She is also a recipient of the NSF Career Award, as well as research awards from Microsoft, Toyota Research, Google, Amazon, JP Morgan, and Sloan Foundation. To learn more about Shuran’s work, please visit: https://www.cs.columbia.edu/~shurans/

Recording link: https://uofi.box.com/s/vrgcq42u5t3avqmwxsf28w27z23wyfov

Abstract:
From autonomous cars in cities to mobile manipulators at home, robots must interact with people. What makes this hard is that human behavior—especially when interacting with other agents—is vastly complex, varying between individuals, environments, and over time. Thus, robots rely on data and machine learning throughout the design process and during deployment to build and refine models of humans. However, by blindly trusting their data-driven human models, today’s robots confidently plan unsafe behaviors around people, resulting in anything from miscoordination to dangerous collisions. My research aims to ensure safety in human-robot interaction, particularly when robots learn from and about people. In this talk, I will discuss how treating robot learning algorithms as dynamical systems driven by human data enables safe human-robot interaction. I will first introduce a Bayesian monitor which infers online if the robot’s learned human model can evolve to well-explain observed human data. I will then discuss how control-theoretic tools enable us to formally quantify what the robot could learn online from human data and how quickly it could learn it. Coupling these ideas with robot motion planning algorithms, I will demonstrate how robots can safely and automatically adapt their behavior based on how trustworthy their learned human models are. I will end this talk by taking a step back and raising the question: “What is the ‘right’ notion of safety when robots interact with people?” and discussing opportunities for how rethinking our notions of safety can capture more subtle aspects of human-robot interaction.

Speaker Bio:
Andrea Bajcsy is a postdoctoral scholar at UC Berkeley in the Electrical Engineering and Computer Science Department and an incoming Assistant Professor at the Robotics Institute at CMU (starting Fall 2023). She studies safe human-robot interaction, particularly when robots learn from and learn about people. Andrea received her Ph.D. in Electrical Engineering & Computer Science from UC Berkeley and B.S. in Computer Science at the University of Maryland, College Park.

Recording link: https://uofi.box.com/s/hlzxrcpbj66npra85tnahnr9ewlzf0gh

Abstract:
In this talk, I discuss our work on using Graph Neural Networks (GNNs) to solve multi-agent coordination problems. I begin by describing how we use GNNs to find a decentralized solution by learning what the agents need to communicate to one another. This communication-based policy is able to achieve near-optimal performance; moreover, when combined with an attention mechanism, we can drastically improve generalization to very-large-scale systems. Next, I consider the inverse problem: instead of optimizing agent policies, what if we could modify the navigation environment, instead? Towards that end, I introduce an environment optimization approach that guarantees the existence of complete solutions, improving agent navigation success rates over heuristic methods. Finally, I discuss challenges in the transfer of learned policies to the real world.

Abstract:
“Amanda Prorok is Professor of Collective Intelligence and Robotics in the Department of Computer Science and Technology at Cambridge University, and a Fellow of Pembroke College. She has been honoured by numerous research awards, including an ERC Starting Grant, an Amazon Research Award, the EPSRC New Investigator Award, the Isaac Newton Trust Early Career Award, and several Best Paper awards. Her PhD thesis was awarded the Asea Brown Boveri (ABB) prize for the best thesis at EPFL in Computer Science. She serves as Associate Editor for IEEE Robotics and Automation Letters (R-AL) and Associate Editor for Autonomous Robots (AURO). Prior to joining Cambridge, Amanda was a postdoctoral researcher at the General Robotics, Automation, Sensing and Perception (GRASP) Laboratory at the University of Pennsylvania, USA, where she worked with Prof. Vijay Kumar. She completed her PhD at EPFL, Switzerland, with Prof. Alcherio Martinoli.”

Recording link: https://uofi.box.com/s/oaij8jo0oon7yye77rpn8x9rx464tx1a

Abstract:
To deploy robots in unstructured, human-centric environments, we must guarantee their ability to safely and reliably complete tasks. In such environments, uncertainty runs rampant and robots invariably need data to refine their autonomy stack. While machine learning can leverage data to obtain components of this stack, e.g., task constraints, dynamics, and perception modules, blindly trusting these potentially unreliable models can compromise safety. Determining how to use these learned components while retaining unified, system-level guarantees on safety and robustness remains an urgent open problem. In this talk, I will present two lines of research towards achieving safe learning-based autonomy. First, I will discuss how to use human task demonstrations to learn hard constraints which must be satisfied to safely complete that task, and how we can guarantee safety by planning with the learned constraints in an uncertainty-aware fashion. Second, I will discuss how to determine where learned perception and dynamics modules can be trusted, and to what extent. We imbue a motion planner with this knowledge to guarantee safe goal reachability when controlling from high-dimensional observations (e.g., images). We demonstrate that these theoretical guarantees translate to empirical success, in simulation and on hardware.

Speaker Bio:
Glen Chou is a postdoctoral associate at MIT CSAIL, advised by Prof. Russ Tedrake. His research focuses on end-to-end safety and robustness guarantees for learning-enabled robots. Previously, Glen received an MS and PhD in Electrical and Computer Engineering from the University of Michigan in 2022, and dual B.S. degrees in Electrical Engineering and Computer Science and Mechanical Engineering from UC Berkeley in 2017. He is a recipient of the National Defense Science and Engineering Graduate (NDSEG) fellowship and is an R:SS Pioneer.

Recording link: https://uofi.box.com/s/wri23ddc58du8fzuhundk5gmbjh36nmj

Abstract:
Collaborative robots coexist with humans in unstructured environments and engage in various tasks involving physical interaction through the entire body. To ensure the safety and versatility of these robots, it is desirable to utilize soft tactile sensors that provide mechanical compliance and tactile data simultaneously. The mechanical property of the soft material effectively mitigates the risk of physical contacts, while the tactile data enable active compliance or social interaction. For this reason, many studies have been conducted to develop soft tactile sensors, but their extension to whole-body robot skin is partially hindered by practical limitations such as low scalability and poor durability. Thus, it is worthwhile to devise an optimal approach to implement whole-body robot skin. In this talk, I will present two works on a soft whole-body robot skin for safe and contact-rich human-robot collaboration. Firstly, I will introduce a biomimetic robot skin that imitates the features of human skin, such as protection and multimodal tactile sensation. Then, I will discuss the methods to implement the biomimetic robot skin (i.e., tomography) and demonstrate its capabilities to sense multi-modal tactile data over a large area. Secondly, a soft pneumatic robot skin will also be presented along with its working principle. This robot skin has a simple structure and functionality but has been seamlessly integrated into the robot arm and used to demonstrate safe and intuitive physical human-robot interaction. Finally, I will examine the significances and limitations of these works and discuss how they can be improved.

Speaker Bio:
Kyungseo Park is a postdoctoral researcher at UIUC Coordinated Science Laboratory, advised by Prof. Joohyung Kim. His research interests include robotics, physical human-robot interaction, and soft robotics. Kyungseo received a B.S., M.S., and Ph.D. in Mechanical Engineering from Korea Advanced Institute for Science and Technology (KAIST), South Korea, in 2016, 2018, and 2022, respectively

Recording link: https://uofi.box.com/s/1vh26s5lpit3s077zbnj0v3opsv08ge9

Abstract:
“Multi-robot teams inherent features of redundancy, increased spatial coverage, flexible reconfigurability, and fusion of distributed sensors and actuators make these systems particularly suitable for applications such as precision agriculture, search-and-rescue operations, or environmental monitoring. In such scenarios, coverage control constitutes an attractive coordination strategy for a swarm, since it allows the robots in a team to spread over a domain according to the importance of its regions: the higher the relevance of an area for the objective of the application, the higher the concentration of robots will be. The coverage paradigm typically considers that all the robots can equally contribute to the coverage task and that the coverage objective is fully known prior deployment of the team. In this talk, we consider realistic scenarios where swarms need to simultaneously monitor multiple types of features (e.g. radiation, humidity, temperature) concurrently at different locations, which require a mixture of sensing capabilities too extensive to be designed into every individual robot. This challenge is addressed by considering heterogeneous multi-robot teams, where each robot is equipped with a subset of those sensors as long as, collectively, the team has all the sensor modalities needed to monitor the collection of features in the domain. Furthermore, we dive into the scenario where robots need to monitor an environment without previous knowledge of its spatial distribution of features. To achieve this, we present an approach where the team simultaneously learns the coverage objectives and optimizes its spatial allocation accordingly, all via local interactions within the team and with the environment. Towards the end of the talk, we move away from the conventional applications of robotic swarms to touch upon how coverage can serve as an interaction modality for artists to effectively utilize robotic swarms for artistic expression. In particular, we focus on the heterogeneous variation of coverage as the means to interactively control desired concentrations of color throughout a canvas for the purpose of artistic multi-robot painting.”

Abstract:
María Santos is a Postdoctoral Research Associate in the Department of Mechanical and Aerospace Engineering at Princeton University, where she works with Dr. Naomi Leonard. María completed her PhD in Electrical and Computer Engineering at the Georgia Institute of Technology in 2020, advised by Dr. Magnus Egerstedt. Prior to that, she received the M.S. degree in Industrial Engineering (Ingeniera Industrial) in 2013 from the University of Vigo, Vigo, Spain and a M.S. degree in Electrical and Computer engineering from the Georgia Institute of Technology, Atlanta, GA, USA, in 2016 as a Fulbright scholar. María’s research deals with the distributed coordination of large multi-agent and multi-robot systems, with a particular focus on modeling heterogeneous teams and the execution of dynamic or unknown tasks. She is also very interested in exploring how to use swarm robotics in various forms of artistic expression, research for which she was awarded a La Caixa Fellowship for Graduate Studies in North America during her doctoral studies.

Recording link: https://uofi.box.com/s/acsn0l0dznb8nz48y8et7hrskyhe8xb4

Abstract:
In recent years, learning-based control paradigms have seen many success stories on various systems and robots. However, as these robots prepare to enter the real world, operating safely in the presence of imperfect model knowledge and external disturbances will be vital to ensure mission success. In the first part of the talk, we present an overview of L1 adaptive control, how it enables safety in autonomous robots, and discuss some of its success stories in the aerospace industry. In the second part of the talk, we present some of our recent results that explore controller tuning using machine learning tools while preserving the controller structure and stability properties. An overview of different projects at our lab that build upon this framework will be demonstrated to show different applications.

Speaker Bio:
Sheng Cheng received the B.Eng. degree in Control Science and Engineering from the Harbin Institute of Technology in China and the M.S. and Ph.D. degrees in Electrical Engineering from the University of Maryland. He is a Postdoctoral Research Associate with the University of Illinois Urbana–Champaign. His current research interests include aerial robotics, learning-enabled control, adaptive control, and distributed parameter systems.

Recording link: https://uofi.box.com/s/8den6asi11r4fzhmxykab4b11jdth5r7

Abstract:
How can biomimetic robots help us in our quest to understand the natural world? And how can examining the evolution and diversity of animal systems help us design better robots? Using a case study of snake-mimicking soft robots, I describe several categories of bio-inspired robotics that serve multiple distinct research goals. By introducing this categorization, I invite us all to consider the many ways in which robotics and biology can serve each other.

Speaker Bio:
Talia Y. Moore is an Assistant Professor of Robotics and of Mechanical Engineering at the University of Michigan, where she is also affiliated with the Department of Ecology and Evolutionary Biology, and the Museum of Zoology. She examines the biomechanics, evolution, and ecology of animals to create bio-inspired robotic systems and uses these physical models to evaluate biological hypotheses.

Recording link: https://uofi.box.com/s/040luxx1hq176kbucc6qobxxg8shmvuh

Abstract:
Reinforcement learning has been a powerful tool for building continuously improving systems in domains like video games and animated character control, but has proven relatively more challenging to apply to problems in real world robotics. In this talk, I will argue that this challenge can be attributed to a mismatch in assumptions between typical RL algorithms and what the real world actually provides, making data collection and utilization difficult. In this talk, I will discuss how to build algorithms and systems to bridge these assumptions and allow robotic learning systems to operate under the assumptions of the real world. In particular, I will describe how we can develop algorithms to ensure easily scalable supervision from humans, perform safe, directed exploration in practical time scales by leveraging prior data and enable uninterrupted autonomous data collection at scale. I will show how these techniques can be applied to real world robotic systems and discuss how these have the potential to be applicable more broadly across a variety of machine learning applications. Lastly, I will provide some perspectives on how this opens the door towards future deployment of robots into unstructured human-centric environments.

Speaker Bio:
Abhishek Gupta is an assistant professor in the Paul G. Allen School for Computer Science and Engineering at the University of Washington. He was formerly a postdoctoral fellow at MIT, working with Pulkit Agrawal and Russ Tedrake. He completed his PhD at UC Berkeley working with Pieter Abbeel and Sergey Levine, building systems that can leverage reinforcement learning algorithms to solve robotics problems. He is interested in research directions that enable directly performing reinforcement learning directly in the real world — reward supervision in reinforcement learning, large scale real world data collection, learning from demonstrations, and multi-task reinforcement learning. He has also spent time at Google Brain and is a recipient of the NDSEG and NSF graduate research fellowships. A more detailed description can be found at https://homes.cs.washington.edu/~abhgupta/

Abstract:
Since the 1960’s we have lived with the promise of one day being able to own a robot that would be able to co-exist, collaborate and cooperate with humans in our everyday lives. This promise has motivated a vast amount of research in the last decades on motion planning, machine learning, perception and physical human-robot interaction (pHRI). Nevertheless, we are yet to see a truly collaborative robot navigating and manipulating objects, the environment or physically collaborating with humans and other robots outside of labs and in the human-centric dynamic spaces we inhabit; i.e., “in-the-wild”. This bottleneck is due to a robot-centric set of assumptions of how humans interact and adapt to technology and machines. In this talk, I will introduce a set of more realistic human-centric assumptions and I posit that for collaborative robots to be truly adopted in such dynamic, ever-changing environments they must possess human-like characteristics of reactivity, compliance, safety, efficiency and transparency. Combining these objectives is challenging as providing a single optimal solution can be intractable and even infeasible due to problem complexity and contradicting goals. Hence, I will present possible avenues to achieve these requirements. I will show that by adopting a Dynamical System (DS) based approach for motion planning we can achieve reactive, safe and provably stable robot behaviors while efficiently teaching the robot complex tasks with a handful of demonstrations. Further, I will show that such an approach can be extended to offer task-level reactivity and can be adopted to efficiently and incrementally learn from failures, as humans do. I will also discuss the role of compliance in collaborative robots, the allowance of soft impacts and the relaxation to the standard definition of safety in pHRI and how this can be achieved with DS-based and optimization-based approaches. I will then talk about the importance of both end-users and designers having a holistic understanding of their robot’s behaviors, capabilities, and limitations and present an approach that uses Bayesian posterior sampling to achieve this. The talk will end with a discussion of open challenges and future directions to achieve truly collaborative robots in-the-wild.

Speaker Bio:
Nadia Figueroa is the Shalini and Rajeev Misra Presidential Assistant Professor in the Mechanical Engineering and Applied Mechanics (MEAM) Department at the University of Pennsylvania. She holds a secondary appointment in the Computer and Information Science (CIS) department and is a faculty advisor at the General Robotics, Automation, Sensing & Perception (GRASP) laboratory. Before joining the faculty, she was a Postdoctoral Associate in the Interactive Robotics Group of the Computer Science and Artificial Intelligence Laboratory (CSAIL) at the Massachusetts Institute of Technology (MIT), advised by Prof. Julie A. Shah. She completed a Ph.D. (2019) in Robotics, Control and Intelligent Systems at the Swiss Federal Institute of Technology in Lausanne (EPFL), advised by Prof. Aude Billard. Prior to this, she was a Research Assistant (2012-2013) at the Engineering Department of New York University Abu Dhabi (NYU-AD) and in the Institute of Robotics and Mechatronics (2011-2012) at the German Aerospace Center (DLR). She holds a B.Sc. degree in Mechatronics (2007) from Monterrey Tech (ITESM-Mexico) and an M.Sc. degree in Automation and Robotics (2012) from the Technical University of Dortmund, Germany.
Her main research interest focuses on developing collaborative human-aware robotic systems: robots that can safely and efficiently interact with humans and other robots in the human-centric dynamic spaces we inhabit. This involves research at the intersection of machine learning, control theory, artificial intelligence, perception, and psychology – with a physical human-robot interaction perspective.

Recording link: TBA

Abstract:
Designing safe and reliable robotic assistance for caregiving is a grand challenge in robotics. A sixth of the United States population is over the age of 65 and in 2014 more than a quarter of the population had a disability. Robotic caregivers could positively benefit society; yet, physical robotic assistance presents several challenges and open research questions relating to teleoperation, active sensing, and autonomous control. In this talk, I will present recent techniques and technology that my group has developed towards addressing core challenges in robotic caregiving. First, I will introduce a head-worn interface that enables people with loss of hand function (due to spinal cord injury or neurodegenerative diseases) to teleoperate assistive mobile manipulators. I will then describe capacitive servoing, a new sensing technique for robotic manipulators to sense the human body and track trajectories along the body. Finally, I will present our recent work in robot learning, including policy learning and dynamics modeling, to perform complex manipulation of deformable garments and blankets around the human body.

Speaker Bio:
Zackory Erickson is an Assistant Professor in The Robotics Institute at Carnegie Mellon University, where he leads the Robotic Caregiving and Human Interaction (RCHI) Lab. His research focuses on developing new robot learning, mobile manipulation, and sensing methods for physical human-robot interaction and healthcare. Zackory received his PhD in Robotics and M.S. in Computer Science from Georgia Tech and B.S. in Computer Science at the University of Wisconsin–La Crosse. His work has won the Best Student Paper Award at ICORR 2019 and a Best Paper in Service Robotics finalist at ICRA 2019.

Abstract:
For robots to perform helpful manual tasks, they must be able to physically interact with the real-world. The ability of robots to grasp and manipulate often depends on the strength and reliability of contact conditions, e.g. friction. In this talk, I will introduce how my lab is developing tools for “messy” or adversarial contact conditions to support the design of more capable systems. Coupled with prototyping and experimental exploration, we generate new systems that better embody desired capabilities. In this talk, I will draw upon recent examples including how we are (1) harnessing fluid flow in soft grippers to improve and monitor grasp state in unique ways, (2) modeling granular interaction forces to support new capabilities in sand, and (3) exploring assistive wearable device topologies for collaborative grasping. Specifically, for compute-and-power-limited robots, I will present a lightweight model selection algorithm that learns when a robot should exploit low-latency on-board computation, or, when highly uncertain, query a more accurate cloud model. Then, I will present a collaborative learning algorithm that allows a diversity of robots to mine their real-time sensory streams for valuable training examples to send to the cloud for model improvement. I will conclude this talk by describing my group’s research efforts to co-design the representation of rich robotic sensory data with networked inference and control tasks for concise, task-relevant representations.

Speaker Bio:
Dr. Hannah Stuart is the Don M. Cunningham Assistant Professor in the Department of Mechanical Engineering at the University of California at Berkeley. She received her BS in Mechanical Engineering at the George Washington University in 2011, and her MS and PhD in Mechanical Engineering at Stanford University in 2013 and 2018, respectively. Recent awards include the NASA Early Career Faculty grant and Johnson & Johnson Women in STEM2D grant.

Abstract:
In the past few years, the robotics industry has been growing rapidly due to the intersection of technology maturity and market demand. This exponential growth has given rise to many advanced multidisciplinary roles within the industry that often require a unique combination of skills in mathematics, physics, software engineering, and algorithms. While traditional curriculum in robotics broadly covers the foundations of the field, it can be quite challenging for new graduates to effectively focus their preparation towards specific industry roles without feeling overwhelmed. The first part of this talk will focus on providing a clear, actionable, and comprehensive roadmap for becoming a robotics engineer based on the most recent roles in the industry. In the second part, the interview structure for such roles as well as effective interviewing strategies will be presented. This talk is targeted towards undergraduate and graduate students in mechanical, electrical, aerospace, and robotics engineering who are looking to enter the robotics industry in the near future.

Speaker Bio:
Ardalan Tajbakhsh is currently a PhD candidate at Carnegie Mellon University where he focuses his research on dynamic multi-agent navigation in unstructured environments for real-world applications like warehouse fulfillment, hospital material transportation, environmental monitoring, and disaster recovery. His background spans a healthy mix of academic research and industry experience in robotics. Prior to his PhD, he was a robotics software engineer at Zebra Technologies where he led the algorithm development efforts for multi-robot coordination in warehouse fulfillment. He has previously held other industry roles at iRobot and Virgin Hyperloop One. Ardalan received an undergraduate degree in Mechanical Engineering with honors from UIUC and a masters in Mechanical Engineering with Robotics concentration from CMU.

Abstract:
Intelligent robot companions have the potential to improve the quality of human life significantly by changing how we live, work, and play. While recent advances in software and hardware opened a new horizon of robotics, state-of-the-art robots are yet far from being blended into our daily lives due to the lack of human-level scene understanding, motion control, safety, and rich interactions. I envision legged robots as intelligent machines beyond simple walking platforms, which can execute a variety of real-world motor tasks in human environments, such as home arrangements, last-mile delivery, and assistive tasks for disabled people. In this talk, I will discuss relevant multi-disciplinary research topics, including deep reinforcement learning, control algorithms, scalable learning pipelines, and sim-to-real techniques.

Speaker Bio:
Sehoon Ha is currently an assistant professor at the Georgia Institute of Technology. Before joining Georgia Tech, he was a research scientist at Google and Disney Research Pittsburgh. He received his Ph.D. degree in Computer Science from the Georgia Institute of Technology. His research interests lie at the intersection between computer graphics and robotics, including physics-based animation, deep reinforcement learning, and computational robot design. His work has been published at top-tier venues including ACM Transactions on Graphics, IEEE Transactions on Robotics, and International Journal of Robotics Research, nominated as the best conference paper (Top 3) in Robotics: Science and Systems, and featured in the popular media press such as IEEE Spectrum, MIT Technology Review, PBS News Hours, and Wired.

Abstract:
Aerial robotics have become ubiquitous, but (like most robots) they still struggle to operate at high speed in unstructured, cramped environments. I will present some of our group’s recent work on pushing vehicles’ capabilities with two distinct approaches. First, I will present algorithmic work aiming to enable motion planning at high speed through unstructured environments, with a specific focus on standard multicopters. The second approach is to modify the vehicle’s physical characteristics, to create a fundamentally different vehicle for whom the problem is easier to the changed physics. Specific designs presented will include a highly collision resilient drone; a passively morphing drone capable of significantly reducing its size; and a preview of a passively morphing system capable of reducing its aerodynamic loads.

Speaker Bio:
Mark W. Mueller is an assistant professor of Mechanical Engineering at UC Berkeley where he leads the High Performance Robotics Laboratory (HiPeRLab). His research focuses on the design and control of aerial robots. He joined the mechanical engineering department at UC Berkeley in September 2016, after spending some time at Verity Studios working on a drone entertainment system, installed in the biggest theater on New York’s broadway. He completed his PhD studies at the ETH Zurich in Switzerland in 2015, and received an MSc there in 2011. He received a bachelors degree in mechanical engineering from the University of Pretoria in South Africa.

Abstract:
The ability of autonomous robot systems to perform reliably and effectively in real-world settings depends on precise understanding of the geometry and semantics of their environment based on streaming sensor observations. This talk will present estimation techniques for sparse object-level mapping, dense surface-level mapping, and distributed multi-robot mapping. The talk will highlight object shape models, octree and Gaussian process surface models, and distributed inference in time-varying graphs.

Speaker Bio:
Nikolay Atanasov is an Assistant Professor of Electrical and Computer Engineering at the University of California San Diego, La Jolla, CA, USA. He obtained a B.S. degree in Electrical Engineering from Trinity College, Hartford, CT, USA in 2008 and M.S. and Ph.D. degrees in Electrical and Systems Engineering from University of Pennsylvania, Philadelphia, PA, USA in 2012 and 2015, respectively. His research focuses on robotics, control theory, and machine learning, applied to active perception problems for mobile robots. He works on probabilistic models that unify geometry and semantics in simultaneous localization and mapping (SLAM) and on optimal control and reinforcement learning of robot motion that minimizes uncertainty in these models. Dr. Atanasov’s work has been recognized by the Joseph and Rosaline Wolf award for the best Ph.D. dissertation in Electrical and Systems Engineering at the University of Pennsylvania in 2015, the best conference paper award at the IEEE International Conference on Robotics and Automation (ICRA) in 2017, and an NSF CAREER award in 2021.

Abstract:
The next generation of robotic systems will be in our homes and workplaces, working in highly unstructured environments and manipulating complex deformables objects. The success of these systems hinges on their ability to reason over the complex dynamics of these objects and carefully control their interactions using the sense of touch. To this end, first I’ll present our recent advances in multimodal neural implicit representations of deformable objects. These representations integrate sight and touch seamlessly to model object deformations and are uniquely well equipped to handle robotic sensing modalities. Second, I’ll present our recent progress on tactile control with high-resolution and highly deformable tactile sensors. Specifically, I’ll discuss our work leveraging the Soft Bubbles to gracefully manipulate tools where we handle high-dimensional tactile signatures and the complex dynamics introduced by the sensor compliance. I’ll end the talk on future directions in tactile controls and deformables and present some of the open challenges in these domains.

Speaker Bio:
Nima Fazeli is an Assistant Professor of Robotics and Assistant Professor of Mechanical Engineering at the University of Michigan and the director of the Manipulation and Machine Intelligence (MMint) Lab. Prof. Fazeli’s primary research interest is enabling intelligent and dexterous robotic manipulation with emphasis on the tight integration of mechanics, perception, controls, learning, and planning. Prof. Fazeli received his PhD from MIT (2019) working with Prof. Alberto Rodriguez, where he also conducted his postdoctoral training. He received his MSc from the University of Maryland at College Park (2014) where he spent most of his time developing models of the human (and, on occasion, swine) arterial tree for cardiovascular disease, diabetes, and cancer diagnoses. His research has been supported by the Rohsenow Fellowship and featured in outlets such as The New York Times, CBS, CNN, and the BBC.

Abstract:
I will discuss our progress in building robotic systems that are agile, dexterous, and real-world-ready in their ability to function in diverse scenarios. The key technical challenge of control in contact-rich problems is addressed using machine learning methods and the results will be illustrated via the following case studies:
(i) a dexterous manipulation system capable of re-orienting novel objects.
(ii) an agile quadruped robot operating on diverse natural terrains.
(iii) a system that only requires a few task demonstrations of an object manipulation task to generalize to new object instances in new configurations.

Speaker Bio:
Pulkit is the Steven and Renee Finn Chair Assistant Professor in the Department of Electrical Engineering and Computer Science at MIT, where he directs the Improbable AI Lab. His research interests span robotics, deep learning, computer vision, and reinforcement learning. His work has received the Best Paper Award at Conference on Robot Learning 2021 and Best Student Paper Award at Conference on Computer Supported Collaborative Learning 2011. He is a recipient of the Sony Faculty Research Award, Salesforce Research Award, Amazon Research Award, a Fulbright fellowship, etc. He received his Ph.D. from UC Berkeley, Bachelors’s degree from IIT Kanpur, where he was awarded the Directors Gold Medal and co-founded SafelyYou Inc.

Abstract:
Robots deployed in the real world will interact with many different humans to perform many different tasks in their lifetime, which makes it difficult (perhaps even impossible) for designers to specify all the aspects that might matter ahead of time. Instead, robots can extract these aspects implicitly when they learn to perform new tasks from their users’ input. The challenge is that this often results in representations which pick up on spurious correlations in the data and fail to capture the human’s representation of what matters for the task, resulting in behaviors that do not generalize to new scenarios. Consequently, the representation, or abstraction, of the tasks the human hopes for the robot to perform may be misaligned with what the robot knows. In my work, I explore ways in which robots can align their representations with those of the humans they interact with so that they can more effectively learn from their input. In this talk I focus on a divide and conquer approach to the robot learning problem: explicitly focus human input on teaching robots good representations before using them for learning downstream tasks. We accomplish this by investigating how robots can reason about the uncertainty in their current representation, explicitly query humans for feature-specific feedback to improve it, then use task-specific input to learn behaviors on top of the new representation.

Speaker Bio:
Andreea Bobu is a Ph.D. candidate at the University of California Berkeley in the Electrical Engineering and Computer Science Department advised by Professor Anca Dragan. Her research focuses on aligning robot and human representations for more seamless interaction between them. In particular, Andreea studies how robots can learn more efficiently from human feedback by explicitly focusing on learning good intermediate human-guided representations before using them for task learning. Prior to her Ph.D. she earned her Bachelor’s degree in Computer Science and Engineering from MIT in 2017. She is the recipient of the Apple AI/ML Ph.D. fellowship, is an R:SS and HRI Pioneer, has won best paper award at HRI 2020, and has worked at NVIDIA Research.

Abstract:
As robots continue to move from controlled environments (e.g., assembly lines) into unstructured ones such as roadways, hospitals, and homes, a key open question for roboticists is how to certify the safety of such systems under the wide range of environmental and perceptual conditions robots can encounter in the wild. In this talk, I will argue for a “risk-aware” approach to robot safety, and present methods for robot manipulation and navigation which account for uncertainty through adaptation and risk-awareness. First, I will present a distributed adaptive controller for collaborative manipulation, which allows a team of robots to adapt to parametric uncertainties to move an unknown rigid body along a desired trajectory in SE(3). In the second half of the talk, we will discuss Neural Radiance Fields (NeRFs), a “neural implicit” scene representation that can be generated using only posed RGB images. I will present our recent work leveraging NeRFs for both visual navigation and manipulation, and show how their probabilistic representation of occupancy/object geometry can be used to enable risk-sensitive planning across a variety of problem domains. I will conclude with some broader thoughts on “risk-awareness” and next directions for enabling safety under perceptual uncertainty.

Speaker Bio:
Preston Culbertson is a postdoctoral scholar in the AMBER Lab at Caltech, researching safe methods for robot planning and control using onboard vision. Preston completed his PhD at Stanford University, working under Prof. Mac Schwager, where his research focused on collaborative manipulation and assembly with teams of robots. In particular, Preston’s research interests include integrating modern techniques for computer vision with methods for robot control and planning that can provide safety guarantees. Preston received the NASA Space Technology Research Fellowship (NSTRF) and the “Best Manipulation Paper” award at ICRA 2018.

Abstract:
Many real-world applications require that robots handle the complexity of multi-party social encounters, e.g., delivery robots may need to navigate through crowds, robots in manufacturing settings may need to coordinate their actions with those of human coworkers, and robots in educational environments may help multiple people practice and improve their skills. How can we enable robots to effectively take part in these social interactions? At first glance, multi-party interactions may be seen as a trivial generalization of one-on-one human-robot interactions, suggesting no special consideration. Unfortunately, this approach is limited in practice because it ignores higher-order effects, like group factors, that often drive human behavior in multi-party Human-Robot Interaction (HRI).
In this talk, I will describe two research directions that we believe are important to advance multi-party HRI. One direction focuses on understanding group dynamics and social group phenomena from an experimental perspective. The other one focuses on leveraging graph state abstractions and structured, data-driven methods for reasoning about individual, interpersonal and group-level factors relevant to these interactions. Examples of these research directions include efforts to motivate prosocial human behavior in HRI, balance human participation in conversations, and improve spatial reasoning for robots in human environments. As part of this talk, I will also describe our recent efforts to scale HRI data collection for early system development and testing via online interactive surveys. We have begun to explore this idea in the context of social robot navigation but, thanks to advances in game development engines, it could be easily applied to other HRI application domains.

Speaker Bio:
Marynel Vázquez is an Assistant Professor in Yale’s Computer Science Department, where she leads the Interactive Machines Group. Her research focuses on Human-Robot Interaction (HRI), especially in multi-party and group settings. Marynel is a recipient of the 2022 NSF CAREER Award and two Amazon Research Awards. Her work has been recognized with nominations to Best Paper awards at HRI 2021, IROS 2018, and RO-MAN 2016, as well as a Best Student Paper award at RO-MAN 2022. Prior to Yale, Marynel was a Post-Doctoral Scholar at the Stanford Vision & Learning Lab and obtained her M.S. and Ph.D. in Robotics from Carnegie Mellon University, where she was a collaborator of Disney Research. Before then, she received her bachelor’s degree in Computer Engineering from Universidad Simón Bolívar in Caracas, Venezuela.

Abstract:
Robotic dexterity stands to be the key challenge to making collaborative robots ubiquitous in the home and industry environments, particularly those that require adaptive systems. The last few decades have produced many solutions in this space that include mechanical transducers (pressure sensors) that while effective usually suffer limitations of the resolution, cross-talk, and limited multi-modal sensing at every point. There are passive, soft sensors that through high friction and form-closure envelope items to be manipulated for stable grasps, and while often effective at securing a grasp, such sensors generally do not provide the dexterity needed to re-grasp, perform finger gaiting or truly quantify the stability of a grasp beyond basic immobilization observed through action. Finally, optical tactile sensors have presented many new avenues for research, with leading designs being GelSight and GelSlim for surface reconstruction and force estimation. While optical tactile sensors stand to be robotics best answer so far to sensing sensitivity that approaches anthropomorphic performance, there is still a noticeable gap in robotics research when it comes to performing manipulation tasks, with end-to-end solutions struggling to extend to new complex manipulation tasks without significant (and often unscalable) training.
In this talk, I will present DenseTact an optical tactile sensor that provides calibrated surface reconstruction and forces for a single fingertip. This calibrated, anthropomorphically inspired fingertip design will allow for modularization of the grasping process and open new avenues of research in robotic manipulation towards collaborative robotic applications.

Speaker Bio:
Monroe Kennedy III is an assistant professor in Mechanical Engineering and courtesy of Computer Science at Stanford University.  Prof. Kennedy is the recipient of the NSF Faculty Early Career Award. He leads the Assistive Robotics and Manipulation laboratory (arm.stanford.edu), which develops collaborative robotic assistants by focusing on combining modeling and control techniques together with machine learning tools. Together, these techniques will improve robotic performance for tasks that are highly dynamic, require dexterity, have considerable complexity, and require human-robot collaboration. He received his Ph.D. in Mechanical Engineering and Applied Mechanics and Masters in Robotics at the University of Pennsylvania and was a member of the GRASP Lab.

Abstract:
Since the 1960’s we have lived with the promise of one day being able to own a robot that would be able to co-exist, collaborate and cooperate with humans in our everyday lives. This promise has motivated a vast amount of research in the last decades on motion planning, machine learning, perception and physical human-robot interaction (pHRI). Nevertheless, we are yet to see a truly collaborative robot navigating and manipulating objects, the environment or physically collaborating with humans and other robots outside of labs and in the human-centric dynamic spaces we inhabit; i.e., “in-the-wild”. This bottleneck is due to a robot-centric set of assumptions of how humans interact and adapt to technology and machines. In this talk, I will introduce a set of more realistic human-centric assumptions and I posit that for collaborative robots to be truly adopted in such dynamic, ever-changing environments they must possess human-like characteristics of reactivity, compliance, safety, efficiency and transparency. Combining these objectives is challenging as providing a single optimal solution can be intractable and even infeasible due to problem complexity and contradicting goals. Hence, I will present possible avenues to achieve these requirements. I will show that by adopting a Dynamical System (DS) based approach for motion planning we can achieve reactive, safe and provably stable robot behaviors while efficiently teaching the robot complex tasks with a handful of demonstrations. Further, I will show that such an approach can be extended to offer task-level reactivity and can be adopted to efficiently and incrementally learn from failures, as humans do. I will also discuss the role of compliance in collaborative robots, the allowance of soft impacts and the relaxation to the standard definition of safety in pHRI and how this can be achieved with DS-based and optimization-based approaches. I will then talk about the importance of both end-users and designers having a holistic understanding of their robot’s behaviors, capabilities, and limitations and present an approach that uses Bayesian posterior sampling to achieve this. The talk will end with a discussion of open challenges and future directions to achieve truly collaborative robots in-the-wild.

Speaker Bio:
Nadia Figueroa is the Shalini and Rajeev Misra Presidential Assistant Professor in the Mechanical Engineering and Applied Mechanics (MEAM) Department at the University of Pennsylvania. She holds a secondary appointment in the Computer and Information Science (CIS) department and is a faculty advisor at the General Robotics, Automation, Sensing & Perception (GRASP) laboratory. Before joining the faculty, she was a Postdoctoral Associate in the Interactive Robotics Group of the Computer Science and Artificial Intelligence Laboratory (CSAIL) at the Massachusetts Institute of Technology (MIT), advised by Prof. Julie A. Shah. She completed a Ph.D. (2019) in Robotics, Control and Intelligent Systems at the Swiss Federal Institute of Technology in Lausanne (EPFL), advised by Prof. Aude Billard. Prior to this, she was a Research Assistant (2012-2013) at the Engineering Department of New York University Abu Dhabi (NYU-AD) and in the Institute of Robotics and Mechatronics (2011-2012) at the German Aerospace Center (DLR). She holds a B.Sc. degree in Mechatronics (2007) from Monterrey Tech (ITESM-Mexico) and an M.Sc. degree in Automation and Robotics (2012) from the Technical University of Dortmund, Germany.
Her main research interest focuses on developing collaborative human-aware robotic systems: robots that can safely and efficiently interact with humans and other robots in the human-centric dynamic spaces we inhabit. This involves research at the intersection of machine learning, control theory, artificial intelligence, perception, and psychology – with a physical human-robot interaction perspective.

Abstract:
To accomplish human- and animal-level agility in robotic systems, we must holistically understand robot hardware, real-time controls, dynamics, perception, and motion planning. Therefore, it is crucial to design control architecture fully considering both hardware and software. In this talk, I will explain our approaches to tackle the challenges in classical control (e.g., bandwidth of feedback control, uncertainty, and robustness) and high-level planning (e.g., step planning, perception, and trajectory optimization), and how the hardware limits are reflected in controller formulation. The tested robots are point-foot bipeds (Hume, Mercury), robots using liquid-cooling viscoelastic actuators (Draco), and quadruped robots using proprioceptive actuators (Mini-Cheetah). I will also present our ongoing research about a new point-foot biped robot (Pat) and a guide dog robot.

Speaker Bio:
Donghyun joined the faculty of the College of Information and Computer Sciences at the University of Massachusetts Amherst as an Assistant Professor in 2021. Before joining UMass, he was a postdoctoral research associate in the Biomimetic Robotics Lab, MIT, from 2019 to 2020. Donghyun was a postdoctoral research associate in the Human-Centered Robotics Lab, the University of Texas at Austin in 2018, where he received his Ph.D. degree in 2017. He holds an MS in Mechanical Engineering from Seoul National University and a BS in Mechanical Engineering from KAIST, Korea. His work on a new viscoelastic liquid-cooled actuator got the best paper award in Transactions on Mechatronics in 2020. His work published in Transactions on Robotics in 2016 was selected as a finalist for the best whole-body control paper and video.