CN103876756A - Lower limb power-assisted exoskeleton robot gait pattern identification method and system - Google Patents

Abstract

The invention provides a gait pattern recognition method and system for a lower limb-assisted exoskeleton robot. The pressure sensor device embedded in the sole of the lower limb-assisted exoskeleton robot detects the wearer's sole pressure information, and the obtained sole pressure information Compare and match with the walking pattern in the human gait database, determine the walking pattern, and identify the motion phase within a gait cycle, and determine the motion phase; perform data processing and analysis, and transmit the matching recognition results to the lower limbs Assist the exoskeleton robot control system as the control basis. The invention provides the walking mode of the wearer of the lower limb assisting exoskeleton and the detailed motion phase in each gait cycle, and provides rich decision-making basis for the control of the lower limb assisting exoskeleton. It can not only provide the wearer's motion information for the lower limb power-assisted exoskeleton, but also be widely used in human gait recognition.

Description

Gait pattern recognition method and system for lower limb assisted exoskeleton robot

technical field

The present invention relates to a gait pattern recognition method and system for a lower limb-assisted exoskeleton robot based on a plantar dynamometer detection system.

Background technique

The lower limb assisted exoskeleton robot combines the advantages of humans and robots, and combines humans and robots into one system to give full play to the human's action-oriented ability and the robot's ability to bear loads. This lower limb power-assisted exoskeleton mechanism can be worn by the wearer, and maintains the same walking movement as the wearer at all times. Like the human exoskeleton, it replaces the human body to bear the external load, so as to achieve the purpose of walking assistance.

In order to alleviate the problem of lower limb injuries caused by heavy loads and long-distance walking; the social problems caused by the aging population; and the lower limb disabilities caused by traffic, earthquake disasters, etc.; It is of great significance to study the lower limb assisted exoskeleton system, and will have a wide range of application prospects.

For the control of the lower limb assisting exoskeleton robot, the most critical factor is to collect the wearer's motion intention and movement trend. Only by obtaining the wearer's motion intention well can the lower limb assisting exoskeleton be controlled so that it can quickly follow The wearer provides effective assistance.

At present, the acquisition of the wearer's movement intention is the detection of EMG and EEG signals. Due to the weak and unstable signals of EMG and EEG signals, the acquisition, processing and analysis of the signals are very complicated. .

The other is to install angle and acceleration sensing devices on the wearer, and detect the wearer's plantar force information to detect motion information, and then control the lower extremity exoskeleton. In the detection of plantar force information, it is very important to design a force measurement system that is easy to wear and can accurately obtain the walking pattern of the human body and the exact motion phase within a gait cycle.

Contents of the invention

Aiming at the problems existing in the control of the lower limb assisted exoskeleton robot and the demand for the recognition of the wearer's walking motion pattern, the present invention provides a human body motion pattern recognition method, through the pressure sensing device embedded in the sole of the robot shoe. The plantar pressure information is detected, the signal conditioning circuit is designed, the data conversion module, the wireless transmission module, etc. are obtained to obtain the plantar pressure information, and the obtained plantar pressure information is compared and matched with the walking pattern in the human gait database to determine its walking pattern and The motion phase in a gait cycle is identified to realize the real-time detection of the motion state of the lower limb assisted exoskeleton robot, which is convenient for the motion control of the robot.

To achieve the above object, the present invention adopts the following technical solutions:

A method for gait pattern recognition of a lower limb assisted exoskeleton robot,

The pressure sensing device embedded in the sole of the lower limb assisting exoskeleton robot detects the wearer's sole pressure information, and the data processing module obtains the sole pressure information through the signal conditioning circuit and the wireless transmission module;

The data processing module compares and matches the obtained plantar pressure information with the walking pattern in the human gait database, judges the walking pattern, and identifies the motion phase within a gait cycle, and judges the motion phase;

Data processing and analysis are carried out, and the results after matching and recognition are transmitted to the control system of the lower limb-assisted exoskeleton robot as a control basis.

Preferably, the database stores characteristic parameters of motion patterns, including plantar force information of the human body in the motion patterns of walking on flat ground, walking on stairs, walking on slopes, jumping, squatting, and running.

Preferably, the analysis yields the phase of motion within a gait cycle:

When the rear heel pressure is less than the preset threshold and the forefoot pressure gradually increases, it means that the wearer is kicking off the ground;

When the pressure on the sole of the front foot and the back heel is less than the preset threshold, it means that the wearer enters the swing leg working mode;

When the rear heel pressure gradually increases, while the forefoot pressure is less than the preset threshold, it means that the wearer enters the heel strike stage;

When the pressure of the rear heel and the forefoot are both greater than the preset threshold, it means that the wearer enters the stage of the supporting leg working mode.

A lower limb assisted exoskeleton robot gait pattern recognition system, which is used to obtain plantar pressure information and identify the lower limb assisted exoskeleton motion pattern, and upload the identified motion phase to the control system of the host computer, and perform lower limb assisted exoskeleton for the control system. Provide basis for assisting exoskeleton motion control;

Including PVDF pressure sensor, signal conditioning circuit, data conversion module, wireless transmission module, and data processing module embedded between the upper insole and the lower insole in the force-measuring shoe;

Real-time detection of the wearer's movement intention through the pressure insole embedded in the foot of the lower limb assisting exoskeleton robot;

The signal conditioning circuit and the wireless transmission module enable the data processing module to obtain plantar pressure information;

The data processing module compares and matches the obtained plantar pressure information with the walking pattern in the human gait database, judges the walking pattern, and identifies the motion phase within a gait cycle, and judges the motion phase;

Data processing and analysis are carried out, and the results after matching and recognition are transmitted to the control system of the lower limb-assisted exoskeleton robot as a control basis.

Preferably, the signal conditioning circuit includes a charge amplification circuit, a low-pass filter circuit, and a voltage amplification circuit;

In the integral circuit, CA3140 with high input impedance is used as the pre-amplifier;

In the low-pass filter circuit, a second-order Butterworth active low-pass filter circuit formed by combining capacitors and resistors with OP07 chip;

In the voltage amplifying circuit, a high-gain operational amplifier UA741 chip is used to provide output short-circuit protection and free operation of blocking.

Preferably, the gait pattern recognition method of a lower limb assisted exoskeleton robot according to claim 4, characterized in that: the PVDF pressure sensor is located in the middle layer of the insole, and the layout in each foot is 5 on the sole of the forefoot and 3 on the rear heel .

Preferably, the gait pattern recognition method of the lower limbs assisting exoskeleton robot as claimed in claim 4, is characterized in that: the data conversion module adopts multi-channel A/D conversion, uses an integrated data acquisition card, and the data acquisition card simulates the input terminal The number of ports is greater than the number of sensing points for the biped.

The beneficial effects of the present invention are: the present invention provides the walking pattern of the wearer of the lower limb assisting exoskeleton and the detailed movement phase in each gait cycle, and provides rich decision-making basis for the control of the lower limb assisting exoskeleton. It can not only provide the wearer's motion information for the lower limb power-assisted exoskeleton, but also be widely used in human gait recognition. The insole-type pressure detection system is embedded in the exoskeleton robot shoe, which has the advantage of being easy to wear. The use of wireless transmission methods frees the wearer from the shackles of traditional physical media such as cables.

Description of drawings

Fig. 1 is the explanatory block diagram of motion pattern recognition in the embodiment;

Fig. 2 is the layout schematic diagram of the pressure sensor based on PVDF in the sole of the foot in the embodiment;

Fig. 3 is a circuit diagram of charge amplification in the signal conditioning circuit;

Fig. 4 is a second-order low-pass filter circuit diagram in the signal conditioning circuit;

Fig. 5 is a voltage amplification circuit diagram in the signal conditioning circuit;

Fig. 6 is a schematic diagram of gait phase analysis of a human body in a gait cycle.

Detailed ways

Preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings.

The present invention is a gait pattern recognition method for a lower-limb-assisted exoskeleton robot, which provides control decisions for a lower-limb-assisted exoskeleton robot, as shown in FIG. 1 .

First of all, the pressure sensing device embedded in the sole of the lower extremity assisting exoskeleton robot detects the wearer's sole pressure confidence, and obtains the sole pressure information through the signal conditioning circuit, data conversion module and wireless transmission module.

The obtained plantar pressure information is matched with the human gait characteristics in the database. The human body gait database stores the plantar force information of the human body in various motion modes such as walking on flat ground, walking on stairs, walking on slopes, jumping, squatting, and running. Based on the matching results, the motion mode of the lower limb-assisted exoskeleton robot is preliminarily judged.

Compare and match the obtained plantar pressure information with the walking pattern in the human gait database to determine its walking pattern, and identify the movement phase within a gait cycle to determine whether it is in the heel-strike stage or the sole-flat support stage , which kind of motion phase in the stage of kicking the toes off the ground and the swing stage.

Carry out further gait recognition in the exercise mode on the acquired information, and compare the preset thresholds of the sensor output pressures of the forefoot and rear heel respectively. When the rear heel pressure is less than the preset threshold and the forefoot pressure gradually increases, it means The wearer is kicking off the ground; when the pressure on the front sole and the rear heel are both lower than the preset threshold, it means that the wearer enters the swing leg working mode stage; The wearer enters the heel strike stage; when the pressure of the rear heel and the forefoot are greater than the preset threshold, it means that the wearer enters the stage of supporting leg work mode.

The identified motion pattern of the lower limb-assisted exoskeleton is transmitted to the host computer control system for motion control of the lower limb-assisted exoskeleton robot.

Among them, the layout of the pressure sensors is shown in Figure 2. There are 5 on the forefoot and 3 on the rear heel, and the pressure sensors on the forefoot or the rear heel are evenly distributed in a triangle. PVDF is fixed in the upper and lower interlayers.

The signal conditioning circuit includes a charge amplification circuit, also known as an integrating circuit, a second-order low-pass filter circuit, and a voltage amplification circuit. Since the output signal of the PVDF piezoelectric film is very weak, the signal must be amplified through the pre-charge amplifier. The charge amplifier is essentially an integrating circuit, that is, the charge output by the sensor is accumulated on the integrating capacitor and then output in the form of voltage, so it is also called Precharge-to-voltage conversion circuit. The charge amplifier circuit is a very sensitive circuit, and it is very likely to introduce interference signals into the circuit. Therefore, it is necessary to filter the signal output by the charge amplifier circuit, and a filter circuit must be provided. Low-temperature drift, high-precision SMD operational amplifiers and resistors and capacitors are used to reduce the size of the circuit board. Between the sensing unit and the circuit board, a multi-core shielded wire is used to shield the external interference signal and reduce the loss of the transmission signal.

As shown in Figure 3, in the integral circuit, CA3140 with high input impedance is selected as the pre-op amplifier. At the same time, the circuit also includes two important components, the feedback capacitor C1 and the feedback resistor R2. In addition, in order to protect the operational amplifier CA3140, a resistor R3 is connected in series at its inverting input terminal, and a capacitor C3 is connected in parallel at both ends of R3 to realize phase compensation. This can prevent R3 and the input capacitance of the operational amplifier CA3140 from forming two poles. The op amp generates self-oscillation.

As shown in Figure 4, in the low-pass filter circuit, a second-order Butterworth active low-pass filter circuit formed by combining capacitors and resistors using the OP07 chip. As shown in Figure 5, in the voltage amplifying circuit, a high-gain operational amplifier UA741 chip is used. This type of monolithic silicon integrated circuit device provides output short-circuit protection and latch-free operation.

The characteristic parameters of various motion modes are stored in the database, including the plantar force information of the human body in various motion modes such as walking on flat ground, walking on stairs, walking on slopes, jumping, squatting, and running. Carry out experimental research on various movement modes, extract the characteristics of the plantar pressure movement, and store them in the database, which is convenient for the matching and comparison of pressure information in the real-time movement of the lower limb assisted exoskeleton and the determination of the walking mode.

In order to carry out accurate gait phase recognition based on the acquired information, Fig. 6 is a movement phase diagram in a gait cycle in level ground walking. It can be seen from the phase diagram that in a gait cycle, the movement of the human body can be It is divided into support phase and swing phase, and further refined into heel strike, full foot flat, mid support phase, heel off the ground, toe off the ground, acceleration push off, swing mid phase, and deceleration landing phase. According to the information of 5 sensors distributed on the forefoot and 3 sensors on the rear heel, the movement phase can be analyzed. For example, when the pressure on the rear heel is less than the preset threshold and the pressure on the forefoot gradually increases, it means that the wearer is kicking off the ground ; When the pressure on the front sole and the rear heel are both lower than the preset threshold, it means that the wearer enters the swing leg working mode; when the pressure on the rear heel gradually increases, and the pressure on the front sole is less than the preset threshold, it means that the wearer enters the heel strike stage ; When the pressure of the rear heel and the forefoot are greater than the preset threshold, it means that the wearer enters the supporting leg working mode stage.

  Upload the identified motion phase to the control system of the host computer to facilitate the implementation of the control strategy for the lower limb assist exoskeleton.

The above-mentioned embodiments are only used to illustrate the present invention, and the selection and layout of the plantar pressure sensor, the design of the signal conditioning circuit, and the identification process of the gait pattern can all be changed. On the basis of the technical solution of the present invention, people in the art Some modifications or alternatives that can be obviously conceived by the skilled person shall fall within the protection scope of the present invention.

Claims (7)

1. A lower limb assisted exoskeleton robot gait pattern recognition method, is characterized in that: The pressure sensing device embedded in the sole of the lower limb assisting exoskeleton robot detects the wearer's sole pressure information, and the data processing module obtains the sole pressure information through the signal conditioning circuit and the wireless transmission module; The data processing module compares and matches the obtained plantar pressure information with the walking pattern in the human gait database, judges the walking pattern, and identifies the motion phase within a gait cycle, and judges the motion phase; Data processing and analysis are carried out, and the results after matching and recognition are transmitted to the control system of the lower limb-assisted exoskeleton robot as a control basis. 2. the lower limbs assisting exoskeleton robot gait pattern recognition method as claimed in claim 1, is characterized in that, deposits the feature parameter of motion pattern in the database, comprises human body flat ground walking, stair walking, slope walking, jumping, squatting down, Plantar force information in running mode. 3. the gait pattern recognition method of the lower limbs assisted exoskeleton robot as claimed in claim 1 or 2, is characterized in that: the motion phase in a gait cycle is identified; When the rear heel pressure is less than the preset threshold and the forefoot pressure gradually increases, it means that the wearer is kicking off the ground; When the pressure on the sole of the front foot and the back heel is less than the preset threshold, it means that the wearer enters the swing leg working mode; When the rear heel pressure gradually increases, while the forefoot pressure is less than the preset threshold, it means that the wearer enters the heel strike stage; When the pressure of the rear heel and the forefoot are both greater than the preset threshold, it means that the wearer enters the stage of the supporting leg working mode. 4. A system for realizing the method according to any one of claims 1-3, which is used to obtain plantar pressure information and identify the lower limb power-assisted exoskeleton movement pattern, and upload the identified movement phase to the control system of the host computer , to provide a basis for the control system to control the movement of the lower extremity assisted exoskeleton; it is characterized in that: Including PVDF pressure sensor, signal conditioning circuit, data conversion module, wireless transmission module, and data processing module embedded between the upper insole and the lower insole in the force-measuring shoe; Real-time detection of the wearer's movement intention through the pressure insole embedded in the foot of the lower limb assisting exoskeleton robot; The signal conditioning circuit and wireless transmission module enable the data processing module to obtain plantar pressure information; The data processing module compares and matches the obtained plantar pressure information with the walking pattern in the human gait database, judges the walking pattern, and identifies the motion phase within a gait cycle, and judges the motion phase; Data processing and analysis are carried out, and the results after matching and recognition are transmitted to the control system of the lower limb-assisted exoskeleton robot as a control basis. 5. the lower limbs assist exoskeleton robot gait pattern recognition system as claimed in claim 4, is characterized in that, signal conditioning circuit comprises charge amplification circuit, low-pass filter circuit, voltage amplification circuit; In the integral circuit, CA3140 with high input impedance is used as the pre-amplifier; In the low-pass filter circuit, a second-order Butterworth active low-pass filter circuit formed by combining capacitors and resistors with OP07 chip; In the voltage amplifying circuit, a high-gain operational amplifier UA741 chip is used to provide output short-circuit protection and free operation of blocking. 6. The gait pattern recognition system for a lower limb assisted exoskeleton robot as claimed in claim 4, wherein the PVDF pressure sensor is located in the middle layer of the insole, and the layout in each foot is 5 on the sole of the forefoot and 3 on the rear heel. 7. The gait pattern recognition system of the exoskeleton robot assisted by lower limbs as claimed in any one of claims 4-6, characterized in that: the data conversion module adopts multi-channel A/D conversion, uses an integrated data acquisition card, and the data acquisition The number of card analog input ports is greater than the number of sensing points of the biped.

Images