CN113397918A - Wearable elbow joint exoskeleton rehabilitation control system - Google Patents

Abstract

The invention discloses a wearable elbow joint exoskeleton rehabilitation control system, which relates to the field of exoskeleton rehabilitation robots. It includes a microprocessor, a wireless communication device and a CAN communication device. The motor module includes a motor driver and a motor; the EMG acquisition module and the attitude acquisition module are connected to the controller module through the wireless communication device, and the motor driver is connected to the controller module through the CAN communication device. , after receiving and analyzing its instructions, it controls the movement of the motor; the training modes include passive training mode, mirror training mode and assisted training mode; the assisted training mode can provide auxiliary torque during the patient's autonomous training to optimize rehabilitation training. The modular structure of the invention can improve the maintainability of the system, and at the same time optimize the training mode, meet the diverse rehabilitation needs of patients, and improve the rehabilitation efficiency.

Description

A wearable elbow joint exoskeleton rehabilitation control system

technical field

The invention relates to the field of exoskeleton robots, in particular to a wearable elbow joint exoskeleton rehabilitation control system.

Background technique

According to the "China Stroke Prevention and Control Report", the number of stroke patients over the age of 40 in my country is as high as 12.42 million, and 55% to 75% of stroke patients will develop hemiplegia, resulting in serious movement disorders and even loss of ability to move. . The existing theory and practice of rehabilitation medicine have proved that correct, scientific and reasonable rehabilitation training can effectively alleviate the development of the disease and play a very important role in restoring the patient's limb function.

Traditional treatment methods mainly rely on medical staff to conduct one-on-one auxiliary training for patients. This training method has high technical requirements for personnel and is time-consuming and labor-intensive. With the continuous development of robotics, people have combined robotics with medical theory and developed various rehabilitation robots to replace human manual operations. Rehabilitation robots not only reduce the burden of medical staff, but also can stably assist patients in rehabilitation training for a long time.

At present, most rehabilitation robots are large and expensive, which is not conducive to community and home rehabilitation, and rehabilitation treatment in large comprehensive rehabilitation hospitals brings a heavy economic burden to patients and their families. Wearable exoskeletons have distinct advantages, light structure and high adaptability, which meet the huge demand for rehabilitation treatment from hospitals to communities and families. The control system of the existing portable upper limb rehabilitation robot is mostly a whole, the control system is relatively complicated, and the maintenance is inconvenient. The patient's active motion intention and poor human-computer interaction make it impossible for the patient to use the exoskeleton rehabilitation robot to achieve an ideal training effect. Therefore, those skilled in the art are committed to developing a wearable, lightweight, multi-rehabilitation training mode elbow exoskeleton rehabilitation control system.

SUMMARY OF THE INVENTION

In view of the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is to optimize the control system of the exoskeleton robot to improve the convenience and maintainability; Multiple options for upper limb rehabilitation on the diseased side to enhance the rehabilitation effect.

In order to achieve the above purpose, the present invention provides a wearable elbow joint exoskeleton rehabilitation control system, which uses a modular concept and a wireless connection method to improve the maintainability of the system; The mode selects different modules to provide accurate data for the training mode, so as to achieve a more ideal training effect.

The present invention is achieved through the following technical solutions: a wearable elbow joint exoskeleton rehabilitation control system, comprising a controller module, a training mode button, an electromyography acquisition module, a posture acquisition module, a motor module and a host computer; wherein,

The controller module includes a microprocessor, a wireless communication device and a CAN communication device, and the motor module includes a motor driver and a motor;

The EMG acquisition module and the posture acquisition module are connected to the controller module through the wireless communication device, and the microprocessor receives the EMG acquisition module and the posture acquisition module through the wireless communication device. The data sent wirelessly is analyzed and calculated, and instructions are generated and sent to the motor module;

The motor driver is connected to the controller module through the CAN communication device, receives instructions from the controller module, parses and controls the movement of the motor, and at the same time the motor driver sends the real-time state of the motor to the controller module;

The training mode buttons are arranged on the controller module, including:

The passive training mode button is used to start the passive training mode: the controller module receives the passive training mode button signal, sends a control command to the motor driver according to a preset program, and controls the motor to drive the exoskeleton to drive the patient. The upper limb on the diseased side performs stable flexion and extension movements; the passive training mode provides three different angular velocities to choose from: high, medium and low;

The mirror image training mode button is used to start the mirror image training mode, the controller module receives the mirror image training mode button signal, collects the angle signal of the upper limb on the healthy side of the patient according to the posture acquisition module, and analyzes the elbow joint angle of the upper limb on the healthy side, controlling the motor module to drive the exoskeleton to drive the patient's upper limb on the diseased side to follow the synchronous flexion and extension movement of the upper limb on the healthy side according to the angle of the elbow joint;

The assistive training mode button is used to start the assistive training mode, the controller module receives the assistive training mode button signal, and calculates the assisting torque required by the upper limb on the diseased side according to the active torque provided by the upper limb on the diseased side of the patient , controlling the motor module to drive the exoskeleton to provide the auxiliary torque, so that the upper limb on the diseased side has enough torque to perform flexion and extension movements.

Further, in the assisted training mode, the controller module transmits the surface EMG signal and the angle signal of the patient's diseased upper limb during flexion and extension movement through the EMG acquisition module and the posture acquisition module, combined with A preset neural network obtains the active torque.

Further, the neural network is obtained by analyzing and training the surface EMG signal and the angle signal when the patient's upper limb on the diseased side performs the target flexion and extension movement; the neural network is pre-input to the controller module.

Further, the neural network has 5 input signals and 1 output signal, and the input signals are respectively biceps EMG, triceps EMG, brachioradialis EMG, elbow joint angle signal and elbow joint angular acceleration signal, the biceps brachii EMG signal, the triceps brachii EMG signal and the brachioradialis EMG signal are preprocessed signals; the output signal is the Active torque.

Further, the value of the assisting torque is multiplied by the active torque and the assisting rate.

Further, the auxiliary rate is the data input to the controller module in advance, and the numerical value can be adjusted according to the actual situation.

Further, the EMG acquisition module includes a dry electrode board, a signal processing board, a first microprocessor and a first wireless communication device, and the surface EMG signals collected by the dry electrode board are processed by the signal processing board and the After the first microprocessor is preprocessed, it is sent to the controller module through the first wireless communication device.

Further, the attitude acquisition module includes an attitude sensor, a second microprocessor and a second wireless communication device, and the angle signal collected by the attitude sensor is analyzed and processed by the second microprocessor, and passed through the second wireless communication device. A communication device is sent to the controller module.

Further, the upper computer is a PC host, which is used to receive, display and save the surface EMG signal and the angle signal sent by the controller module, and can also be used to debug the motor module.

Further, the controller module, the myoelectric acquisition module, the attitude acquisition module and the motor module are powered by lithium batteries.

The beneficial effects of the invention are as follows: a wearable elbow joint exoskeleton rehabilitation control system is provided, the structure of the control system is optimized, and the connection of the control system is simplified; at the same time, diversified training modes are provided, which is convenient to use and helps For the recovery of the muscle strength of the patient's upper limb on the diseased side, especially the assistive training mode, a neural network suitable for the patient's individual is introduced. While the patient actively trains the upper limb on the diseased side, it provides auxiliary torque for the diseased side that is beneficial to rehabilitation. The patient's rehabilitation exercise is optimized and the rehabilitation efficiency is improved.

The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings, so as to fully understand the purpose, characteristics and effects of the present invention.

Description of drawings

Fig. 1 is the schematic diagram of the control system of the present invention;

Fig. 2 is the working flow chart of the control system of the present invention;

Fig. 3 is the working flow chart of the assisting training mode in the control system of the present invention;

Fig. 4 is the flow chart of using genetic algorithm to optimize neural network training in the control system of the present invention;

Fig. 5 is an exoskeleton model diagram of the control system of the present invention.

Among them, 1-controller module, 2-myoelectric acquisition module, 3-attitude acquisition module, 4-motor module, 5-host computer, 11-wireless communication device, 12-CAN communication device, 41-motor driver, 42- motor.

Detailed ways

The following describes several preferred embodiments of the present invention with reference to the accompanying drawings, so as to make its technical content clearer and easier to understand. The present invention can be embodied in many different forms of embodiments, and the protection scope of the present invention is not limited to the embodiments mentioned herein.

In the drawings, structurally identical components are denoted by the same numerals, and structurally or functionally similar components are denoted by like numerals throughout. The size and thickness of each component shown in the drawings are arbitrarily shown, and the present invention does not limit the size and thickness of each component. In order to make the illustration clearer, the thicknesses of components are appropriately exaggerated in some places in the drawings.

As shown in FIG. 1 , the present invention provides a wearable elbow joint exoskeleton rehabilitation control system, including a controller module 1, an EMG acquisition module 2, a posture acquisition module 3, a motor module 4, a host computer 5 and a training mode button .

The controller module 1 includes a microprocessor, a wireless communication device 11 and a CAN communication device 12; the wireless communication device 11 receives the data wirelessly sent by the myoelectric acquisition module 2 and the attitude acquisition module 3, analyzes and calculates, and passes the generated instruction through CAN The communication device 12 sends it to the motor module 4 .

The motor module 4 includes a motor driver 41 and a motor 42; the motor driver 41 is connected to the controller module 1 through the CAN communication device 12, receives the instructions of the controller module 1, parses and controls the movement of the motor 42, and at the same time the motor driver 4 connects the motor 42. The real-time status is sent to the controller module 1.

The EMG acquisition module 2 includes a dry electrode board, a signal processing board, a first microprocessor and a first wireless communication device. The dry electrode board conducts EMG signals on the surface of the human skin, and is input to the signal processing board in a differential manner; The board integrates the functions of amplification, filtering and 50Hz notch, which can preprocess the surface EMG signal; the first microprocessor converts the preprocessed surface EMG signal (analog voltage signal) into a digital signal, and communicates through the first wireless communication The device is sent to the controller module 1.

The attitude acquisition module 3 includes an attitude sensor, a second microprocessor and a second wireless communication device. The angle signal collected by the attitude sensor is analyzed and processed by the second microprocessor and sent to the controller module 1 through the second wireless communication device.

The upper computer 5 is a PC host, which is used to receive, display and save the surface EMG signal and the angle signal sent by the controller module 1 , and can also be used to debug the motor module 4 .

The controller module 1, the electromyography acquisition module 2, the attitude acquisition module 3 and the motor module 4 are powered by lithium batteries.

The training mode buttons are arranged on the controller module 1, including: a passive training mode button, used to start the passive training mode; a mirror training mode button, used to start the mirror training mode; an assistive training mode button, used to start the assistive training mode .

As shown in Figure 2, the work flow of the wearable elbow joint exoskeleton rehabilitation control system provided by the present invention is:

Step 1: The system is powered on for initialization. The initialization includes:

(1) The EMG acquisition module 2 and the attitude acquisition module 3 establish a wireless connection with the controller module 1;

(2) The motor 42 automatically finds the position of the absolute zero point.

Step 2: The patient touches the training mode button on the controller module 1 according to the rehabilitation needs, and the controller judges the training mode and works according to the following three training modes:

(1) Passive training mode: after the patient's upper limb on the diseased side wears the exoskeleton, the controller module 1 sends a control command to the motor driver 41 according to the preset program, and then controls the motor 42 to drive the exoskeleton to drive the upper limb on the diseased side to perform stable flexion and extension movements ;The passive training mode can choose three different angular velocities: high, medium and low; the maximum flexion and extension angle of the exoskeleton elbow joint is smaller than the maximum flexion and extension angle of the human upper limb;

(2) Mirror training mode: After the patient wears the exoskeleton on the upper limb on the diseased side, the two posture acquisition modules 3 are respectively fixed on the upper arm and the lower arm of the healthy side upper limb, and the posture acquisition module 3 collects the angle signal of the healthy upper limb and transmits it wirelessly. To the controller module 1; the controller module 1 obtains the elbow joint angle of the healthy side through analysis and calculation, generates a control command and sends it to the motor driver 41; the motor driver 41 controls the motor 42 to drive the exoskeleton on the diseased side to follow the healthy side according to the command. The movement of the elbow joint angle; during the training process, the exoskeleton movement has the maximum movement angular velocity and angle limit, so as to avoid the injury of the upper limb on the diseased side caused by the movement of the upper limb on the healthy side too fast or the angle is too large;

(3) Assisted training mode: As shown in Figure 3, before training, three EMG acquisition modules 2 were fixed on the biceps brachii, triceps brachii and brachioradialis of the upper limb of the patient, respectively. Then, the two posture acquisition modules 3 are respectively fixed on the upper arm and the lower arm of the upper limb on the diseased side, and then the medical staff or family members assist the patient to do flexion and extension of the upper limb on the diseased side, and at the same time, the electromyography acquisition module 2 collects the biceps. The EMG signals of the muscle, triceps brachii and brachioradialis, and the attitude acquisition module 3 collects the angle signals of the upper arm and the lower arm; the collected data is used for offline neural network training to obtain a trained neural network.

When the patient performs assisted movement training, the three myoelectric acquisition modules 2 and the two posture acquisition modules 3 are respectively fixed at the same position above the patient's upper limb on the diseased side, and then the upper limb on the diseased side wears the exoskeleton; During the flexion and extension of the lateral upper limb, the biceps brachii, triceps brachii and brachioradialis muscle related to the flexion and extension exercise will contract or relax, resulting in changes in the surface EMG signal. The EMG acquisition module 2 collects the surface EMG in real time. signal, and wirelessly transmit the data to the controller module 1; the attitude acquisition module 3 collects the angle signal of the upper limb on the diseased side in real time, and wirelessly transmits the data to the controller module 1; the controller module 1 receives the surface EMG signal and the angle After the signal, it is preprocessed, including feature extraction and normalization, etc., to calculate the elbow joint angle and angular velocity; then the three preprocessed surface EMG signals, elbow joint angle and angular velocity, a total of 5 signals Input it into the trained neural network, the output signal of the neural network is the active torque generated by the muscles when the patient flexes and stretches; after obtaining the active torque, according to the set assist rate, the assist rate can be calculated by multiplying the active torque by the assist rate. Then the controller module 1 sends a control command to the motor driver 41, and controls the motor 42 to drive the exoskeleton to drive the upper limb on the diseased side to perform flexion and extension movements according to the auxiliary torque; the size of the auxiliary rate can be adjusted according to the state of the diseased side. Therefore, the active torque of the patient plus the auxiliary torque of the exoskeleton can make the upper limb on the diseased side perform reasonable flexion and extension movements.

The above neural network training process is shown in Figure 4, which is optimized according to the genetic algorithm. Because the neural network randomly initializes the weights and thresholds, and adjusts the weights and thresholds by gradient descent, it is possible for the neural network to have a local optimal solution due to random initialization. Therefore, the genetic algorithm is used to optimize the initial value and threshold of the neural network. . First initialize the weights and thresholds, and then input them into the genetic algorithm for optimization, and then obtain the optimal weights and thresholds, and then train the neural network.

Step 3: The training is completed, the work is finished, and the exoskeleton and the corresponding acquisition module are removed.

During the training process, you can perform the reset operation, and then re-select the training mode.

As shown in FIG. 5 , the three-dimensional view of the exoskeleton of the wearable elbow joint exoskeleton rehabilitation control system provided by the present invention, the overall weight of the exoskeleton is light and easy to wear.

The preferred embodiments of the present invention have been described in detail above. It should be understood that many modifications and changes can be made according to the concept of the present invention by those skilled in the art without creative efforts. Therefore, all technical solutions that can be obtained by those skilled in the art through logical analysis, reasoning or limited experiments on the basis of the prior art according to the concept of the present invention shall fall within the protection scope determined by the claims.

Claims (10)

1. A wearable elbow joint exoskeleton rehabilitation control system is characterized in that, comprising a controller module (1), a training mode button, an electromyography acquisition module (2), a posture acquisition module (3), a motor module (4) and the host computer (5); wherein, The controller module (1) includes a microprocessor, a wireless communication device (11) and a CAN communication device (12), and the motor module (4) includes a motor driver (41) and a motor (42); The myoelectric acquisition module (2) and the attitude acquisition module (3) are connected with the controller module (1) through the wireless communication device (11), and the microprocessor is connected through the wireless communication device (11). 11) Receive data wirelessly sent by the myoelectric acquisition module (2) and the attitude acquisition module (3), analyze and calculate and generate an instruction, and send it to the motor module (4); The motor driver (41) is connected to the controller module (1) through the CAN communication device (12), receives an instruction from the controller module (1), parses and controls the motor (42) movement, while the motor driver (41) sends the real-time state of the motor (42) to the controller module (1); The training mode buttons are arranged on the controller module (1), including: Passive training mode button, used to start the passive training mode: the controller module (1) receives the passive training mode button signal, sends a control command to the motor driver (41) according to a preset program, and controls the motor (42) driving the exoskeleton to drive the patient's upper limb on the diseased side to perform stable flexion and extension movements; the passive training mode provides three different angular velocities of high, medium and low for selection; The mirror image training mode button is used to start the mirror image training mode, the controller module (1) receives the mirror image training mode button signal, collects the angle signal of the upper limb on the healthy side of the patient according to the posture acquisition module (3), and analyzes the healthy side The elbow joint angle of the upper limb is controlled, and the motor module (4) is controlled to drive the exoskeleton to drive the patient's upper limb on the diseased side to perform synchronous flexion and extension movements following the upper limb on the healthy side according to the elbow joint angle; The assistive training mode button is used to start the assistive training mode, and the controller module (1) receives the assistive training mode button signal, and calculates the required upper limb on the affected side according to the active torque provided by the patient's affected upper limb The auxiliary torque is controlled, and the motor module (4) is controlled to drive the exoskeleton to provide the auxiliary torque, so that the upper limb on the diseased side has sufficient torque to perform flexion and extension movements. 2. The wearable elbow joint exoskeleton rehabilitation control system according to claim 1, wherein the controller module (1) is in the assisted training mode, through the myoelectric acquisition module (2) and the posture The surface electromyographic signal and the angle signal of the patient's diseased upper limb during flexion and extension movements transmitted by the acquisition module (3) are combined with a preset neural network to obtain the active torque. 3. The wearable elbow joint exoskeleton rehabilitation control system as claimed in claim 2, wherein the neural network is obtained by analyzing and training the surface EMG signal and the angle signal when the patient's upper limb on the diseased side does target flexion and extension exercise; A neural network is pre-input to the controller module (1). 4. The wearable elbow joint exoskeleton rehabilitation control system as claimed in claim 3, wherein the neural network has 5 input signals and 1 output signal, and the input signals are respectively biceps brachii electromyogram, triceps brachii The EMG signal of the head muscle, the EMG signal of the brachioradialis muscle, the angle signal of the elbow joint and the angular acceleration signal of the elbow joint, the EMG signal of the biceps muscle, the EMG signal of the triceps muscle and the brachioradialis muscle The electrical signal is a preprocessed signal; the output signal is the active torque. 5 . The wearable elbow joint exoskeleton rehabilitation control system according to claim 1 , wherein the value of the assisting moment is multiplied by the active moment and the assisting rate. 6 . 6. The wearable elbow joint exoskeleton rehabilitation control system according to claim 5, wherein the auxiliary rate is pre-inputted into the controller module (1), and the value is adjusted according to the actual situation. 7. The wearable elbow joint exoskeleton rehabilitation control system according to claim 1, wherein the electromyography acquisition module (2) comprises a dry electrode board, a signal processing board, a first microprocessor and a first wireless communication device, so The surface EMG signal collected by the dry electrode plate is preprocessed by the signal processing plate and the first microprocessor, and then sent to the controller module (1) through the first wireless communication device. 8. The wearable elbow joint exoskeleton rehabilitation control system according to claim 1, wherein the attitude acquisition module (3) comprises an attitude sensor, a second microprocessor and a second wireless communication device, and the angle collected by the attitude sensor The signal is analyzed and processed by the second microprocessor and sent to the controller module (1) through the second wireless communication device. 9. The wearable elbow joint exoskeleton rehabilitation control system according to claim 1, wherein the upper computer (5) is a PC host, used for receiving, displaying and saving the surface sent by the controller module (1) The myoelectric signal and the angle signal are also used to debug the motor module (4). 10. The wearable elbow joint exoskeleton rehabilitation control system according to claim 1, wherein the controller module (1), the electromyography acquisition module (2), the posture acquisition module (3) and the motor The module (4) is powered by a lithium battery.

Images