Exoskeleton robot control method, system, controller, and storage medium

Abstract

The application provides an exoskeleton robot control method, system, controller and storage medium, and relates to the technical field of exoskeleton robots. The method comprises the following steps: obtaining a deformation variable of a human muscle group; inputting the deformation variable into an intention recognition model to obtain an action intention of the human body output by the intention recognition model; inputting the deformation variable into a muscle force mapping model to obtain muscle force of the muscle group output by the muscle force mapping model; determining a control instruction of each joint of the exoskeleton robot based on the action intention and the muscle force of the muscle group; wherein the intention recognition model is obtained by training based on sample data of the deformation variable of the human muscle group and sample data of the action intention of the human body; and the muscle force mapping model is obtained by training based on sample data of the deformation variable of the human muscle group and sample data of the muscle force of the muscle group. In the control process, both the active force provided by the muscle force of the human muscle group and the action intention of the human body are considered, so that the exoskeleton robot adjusts the provided auxiliary force according to the high and low active force of the human body to automatically adapt to the change of the muscle force of the user.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of exoskeleton robots, and in particular to an exoskeleton robot control method, system, controller and storage medium. BACKGROUND

[0002] An exoskeleton robot can be worn on the limbs of an operator to provide rehabilitation training, physical assistance and the like for the operator. Taking a weak patient as an example, the patient can walk and stand like a normal person under the support of the exoskeleton robot after wearing the exoskeleton robot.

[0003] Exoskeleton robot operation needs to obtain the action intention of the operator by means of sensing technology. The common way at present is to detect by means of an electromyography sensor to further determine the action intention of the operator. The electromyography sensor is easily affected by uncertain factors such as human sweat and body temperature, needs to be recalibrated every time it is started, has poor consistency, and the electromyography sensor is difficult to detect signals for the weak muscle group, so the applicable population is limited. SUMMARY

[0004] The present application provides an exoskeleton robot control method, system, controller and storage medium to solve the defects of poor consistency and limited applicable population of the exoskeleton robot using an electromyography sensor in the prior art.

[0005] The present application provides an exoskeleton robot control method, comprising:

[0006] obtaining a deformation variable of a human muscle group;

[0007] inputting the deformation variable into an intention recognition model to obtain an action intention of the human output by the intention recognition model;

[0008] inputting the deformation variable into a muscle force mapping model to obtain muscle force of the muscle group output by the muscle force mapping model;

[0009] determining a control instruction of each joint of an exoskeleton robot based on the action intention and the muscle force of the muscle group;

[0010] The intention recognition model is trained based on deformation variable sample data of a human muscle group and action intention sample data of the human, and the muscle force mapping model is trained based on deformation variable sample data of a human muscle group and muscle force sample data of the muscle group.

[0011] According to the exoskeleton robot control method provided by the present application, before obtaining the deformation variable of the human muscle group, it further comprises:

[0012] obtaining a transmission signal and a return signal of a plurality of ultrasonic sensors;

[0013] determining the deformation variable of the human muscle group based on the phase difference between the transmission signal and the return signal.

[0014] The plurality of ultrasonic sensors are arranged at different positions of the human body.

[0015] The control method of the exoskeleton robot provided by the application comprises the following steps:

[0016] The coupling state of the human body and the exoskeleton robot is determined based on the action intention and a walking dynamics model.

[0017] The output torque of each joint of the human body is determined based on the muscle force.

[0018] The compensation torque of each joint of the exoskeleton robot is determined based on the resultant torque and the output torque.

[0019] The control method of the exoskeleton robot provided by the application comprises the following steps:

[0020] The action recognition model is input with the deformation variables to obtain human actions output by the action recognition model.

[0021] The intention mapping model is input with the human actions to obtain the action intention output by the intention mapping model.

[0022] The action recognition model is trained based on deformation variable sample data of the muscle groups of the human body and human action sample data. The intention mapping model is trained based on human action sample data and action intention sample data.

[0023] The application further provides an exoskeleton robot system comprising an exoskeleton robot and a bandage.

[0024] The plurality of ultrasonic sensors are arranged at different positions of the human body.

[0025] According to the exoskeleton robot system provided by the application, the ultrasonic sensor comprises a host computer, a main control chip, a high-voltage pulse transmitting chip and an echo receiving module, the main control chip is in communication connection with the host computer, the main control chip is used for controlling the working mode and pulse length of the high-voltage pulse transmitting chip, the echo receiving module is used for receiving the echo signal of the ultrasonic wave, and the echo receiving module is in communication connection with the main control chip.

[0026] According to the exoskeleton robot system provided by the application, the ultrasonic sensor further comprises a transmitting-receiving isolation module, which is arranged between the high-voltage pulse transmitting chip and the echo receiving module.

[0027] The application further provides an electronic device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, and the processor implements the steps of the exoskeleton robot control method according to any one of the above when executing the program.

[0028] The application further provides a non-transitory computer readable storage medium, which stores a computer program, and the computer program implements the steps of the exoskeleton robot control method according to any one of the above when executed by a processor.

[0029] The exoskeleton robot control method, system, controller and storage medium provided by the application obtain the deformation amount of the muscle group of the human body, input the deformation amount into an intention recognition model and a muscle force mapping model respectively, determine the action intention and muscle force of the human body, and then determine the control instruction of the exoskeleton robot, so that the exoskeleton robot adapts to the gait of the human body, and in the control process, both the active force provided by the muscle force of the human body and the action intention of the human body are considered, so that the exoskeleton robot adjusts the auxiliary force provided according to the active force of the human body, to automatically adapt to the change of the muscle force of the user. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the application or prior art, the following will briefly introduce the drawings needed in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can also be obtained by those skilled in the art without creative effort.

[0031] Figure 1 is one of the flowcharts of the exoskeleton robot control method provided by the application;

[0032] Figure 2 is the second flowchart of the exoskeleton robot control method provided by the application;

[0033] Figure 3Figure 3 is a flow diagram of a third embodiment of the exoskeleton robot control method provided by the present application;

[0034] Figure 4 Figure 4 is a structural diagram of an exoskeleton robot provided by the present application;

[0035] Figure 5 Figure 5 is a partial structural diagram of a strap provided by the present application;

[0036] Figure 6 Figure 6 is a structural diagram of a controller provided by the present application. BRIEF DESCRIPTION OF DRAWINGS

[0038] 10, strap; 11, ultrasonic sensor; 12, clamping groove; 13, clamping block; 20, support; 21, wheel; 31, waist and back module; 32, thigh module; 33, calf module; 34, foot bottom module; 41, hip joint torque sensor; 42, knee joint torque sensor; 43, ankle joint torque sensor; 44, hip joint encoder; 45, knee joint encoder; 46, ankle joint encoder; 50, controller. DETAILED DESCRIPTION

[0039] In order to make the objectives, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below with reference to the drawings in the present application. Obviously, the described embodiments are some embodiments of the present application, but not all embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0040] The exoskeleton robot control method provided by the present application will be described below with reference to the drawings. Figures 1-3 The exoskeleton robot control method provided by the present application will be described below with reference to the drawings.

[0041] As shown in Figure 1 , the present application provides an exoskeleton robot control method, which comprises:

[0042] Step 110, obtaining a deformation variable of a human muscle group;

[0043] Step 120, inputting the deformation variable into an intention recognition model to obtain an action intention of the human body output by the intention recognition model; and inputting the deformation variable into a muscle force mapping model to obtain a muscle force of the muscle group output by the muscle force mapping model;

[0044] Step 130, determining a control instruction of each joint of the exoskeleton robot based on the action intention and the muscle force of the muscle group.

[0045] The action intention recognition model is trained by taking the deformation variable sample data of the human muscle group as a sample and taking the action intention of the human body corresponding to the deformation variable as a label.

[0046] The ultrasonic sensor collects the transmission signal and the return signal of the human muscle group, determines the deformation variable of the human muscle group based on the phase difference between the transmission signal and the return signal, and inputs the deformation variable into the action intention recognition model and the muscle force mapping model to determine the action intention of the human body and the muscle force of the muscle group.

[0047] In an embodiment of the present application, the deformation variable of the human muscle group is input into the action intention model to obtain the action category of the human body output by the action intention model; and the action category of the human body is input into the intention instruction mapping model to obtain the action intention of the human muscle group output by the intention instruction mapping model. The action intention includes the target position and the target angle of the human muscle group. In another embodiment of the present application, the deformation variable of the human muscle group is input into the action intention recognition model to directly determine the action intention of the human body through the action intention recognition model.

[0048] Specifically, the action intention is the target action of the human body, the resultant moment of each joint in the human-machine coupling state can be determined based on the action intention, the output moment of each joint of the current human body can be determined based on the muscle force of the muscle group, the complementary moment of each joint of the exoskeleton robot is determined according to the resultant moment of each joint and the output moment of each joint of the current human body, and then the control instruction of each joint of the exoskeleton robot is formed to control the movement of each joint of the exoskeleton robot.

[0049] The exoskeleton robot control method provided by the present application acquires the deformation variable of the human muscle group, inputs it into the action intention recognition model and the muscle force mapping model respectively to determine the action intention of the human body and the muscle force of the muscle group, and then determines the control instruction of the exoskeleton robot, and the exoskeleton robot executes the control instruction to adapt to the gait of the human body.

[0050] In an embodiment of the present application, as shown in Figure 2 Before acquiring the deformation variable of the human muscle group, the method further includes:

[0051] Step 210, acquiring the transmission signal and the return signal of the plurality of ultrasonic sensors;

[0052] Step 220, determining the deformation variable of the human muscle group based on the phase difference between the transmission signal and the return signal. The plurality of ultrasonic sensors are arranged at different positions of the human body.

[0053] Specifically, the ultrasonic sensor collects a transmission signal and a return signal of an ultrasonic wave, and multiple ultrasonic sensors detect from different directions of the human body, and due to the phase difference of the ultrasonic signals in the continuous time sequence in the transmission and reflection process of the ultrasonic wave in the deep muscle tissue of the human body, the all-directional dynamic muscle deformation variable is obtained under the superposition of the ultrasonic signals in multiple directions.

[0054] The control method of the exoskeleton robot provided in the embodiment of the application determines the deformation variable of the muscle group of the human body through the phase difference between the transmission signal and the return signal of the ultrasonic sensor, and compared with the traditional electromyography detection sensor, the ultrasonic sensor can penetrate into the deep muscle of the human body, has higher reliability, and is also applicable to the population with weaker muscle strength; in addition, compared with the traditional electromyography detection sensor, the ultrasonic sensor has higher signal consistency and higher practicability. The traditional electromyography detection sensor is complicated to calibrate, and the ultrasonic sensor has high man-machine fault tolerance and high numerical consistency.

[0055] In a specific embodiment of the application, as shown in Figure 3 The control instruction of each joint of the exoskeleton robot based on the action intention and the muscle strength of the muscle group specifically includes:

[0056] In step 310, the resultant moment of each joint under the man-machine coupling state is determined based on the action intention and the walking dynamics model, and the output moment of each joint of the human body is determined based on the muscle strength of the muscle group.

[0057] In step 320, the compensation moment of each joint of the exoskeleton robot is determined based on the resultant moment and the output moment.

[0058] In step 330, the motion of each joint of the exoskeleton robot is controlled based on the compensation moment.

[0059] The walking dynamics model represents the corresponding relationship between the moments of each joint of the exoskeleton robot under the standard gait of the human body and under the man-machine coupling state.

[0060] The walking gait of a healthy person is collected by using multiple cameras to determine the motion information of key points such as the hip, knee, ankle and foot of the lower limbs of the human body, and the standard gait is obtained after gait segmentation and splicing, median filtering and normalization processing. The action intention is determined based on the intention recognition model, the standard gait to be performed by the human body is determined based on the action intention, the angle, position and other motion information of each joint of the human body are inversely kinematically backstepped, and the moment at each joint of the exoskeleton when the exoskeleton realizes the action intention is determined through foot pressure analysis. It should be noted that the standard action to be performed by the human body in the next step is determined based on the action intention and the standard gait of the human body, the standard action includes the angle information and position information of each joint of the human body, and the moment size of each joint of the exoskeleton when the exoskeleton realizes the standard action is determined based on the standard action.

[0061] When a human body wears an exoskeleton to walk, the muscle can provide part of the walking force, based on the muscle force of the muscle group, the output torque of each joint of the current human body can be determined through kinematic analysis. The compensation torque of the exoskeleton is determined through the resultant torque and the output torque, and thus the assisting force provided by the exoskeleton robot and the force provided by the human body are consistent with the force required by the human body to walk according to the standard gait.

[0062] The exoskeleton robot control method provided by the embodiment of the application determines the resultant torque of each joint of the human body in the human-machine coupling state through the motion intention and the walking dynamics model, determines the output torque of each joint of the human body corresponding to the muscle force of the muscle group through calculation, and thus calculates the compensation torque required by the exoskeleton robot to realize accurate torque compensation control. Compared with the traditional gait control, the exoskeleton provides appropriate compensation torque while considering the force provided by the human body, and better adapts to the current physical state of the user to realize standard gait walking.

[0063] The application further provides an exoskeleton robot control device, which comprises:

[0064] The acquisition unit is used to acquire the deformation variable of the muscle group of the human body.

[0065] The control unit is used to input the deformation variable into the intention recognition model to obtain the motion intention of the human body output by the intention recognition model, input the deformation variable into the muscle force mapping model to obtain the muscle force of the muscle group output by the muscle force mapping model, and determine the control instruction of each joint of the exoskeleton robot based on the motion intention and the muscle force of the muscle group. The intention recognition model is obtained by training with the deformation variable sample data of the muscle group of the human body as samples and the motion intention of the human body corresponding to the deformation variable as labels. The muscle force mapping model is obtained by training with the deformation variable sample data of the muscle group of the human body as samples and the muscle force corresponding to the deformation variable sample data as labels.

[0066] Further, the exoskeleton robot control device further comprises:

[0067] The acquisition unit is further used to acquire the transmission signal and the return signal of a plurality of ultrasonic sensors.

[0068] The processing unit is used to determine the deformation variable of the muscle group of the human body based on the phase difference between the transmission signal and the return signal, and the plurality of ultrasonic sensors are arranged at different positions of the human body.

[0069] In the embodiment of the application, the control unit further comprises: the control unit is further used to determine the resultant torque of each joint in the human-machine coupling state based on the motion intention and the walking dynamics model; and the walking dynamics model represents the corresponding relationship between the standard gait of the human body and the torque of each joint of the exoskeleton robot in the human-machine coupling state.

[0070] The output torque of each joint in the human body is determined based on the muscle strength of the aforementioned muscle groups.

[0071] The compensation torque of each joint of the exoskeleton robot is determined based on the resultant torque and the output torque, and the movement of each joint of the exoskeleton robot is controlled based on the compensation torque.

[0072] In a specific embodiment of the present invention, the control unit is specifically used to input the deformation into the action recognition model to obtain the human body action output by the action recognition model;

[0073] The human body action is input into the intention mapping model to obtain the action intention output by the intention mapping model;

[0074] The action recognition model is trained based on human muscle group deformation sample data and human action sample data; the intention mapping model is trained based on human action sample data and action intention sample data.

[0075] This invention also provides an exoskeleton robot system, such as... Figure 4 As shown, it includes an exoskeleton robot and a strap 10. The exoskeleton robot is equipped with a controller 50, and the strap 10 is equipped with multiple ultrasonic sensors 11. Torque sensors are installed at the joints of the exoskeleton robot. The torque sensors and ultrasonic sensors 11 are all communicatively connected to the controller 50.

[0076] Optionally, multiple ultrasonic sensors 11 are evenly distributed along the circumference of the strap 10.

[0077] The strap 10 includes a first strap body and a second strap body. One end of each strap body has a slot 12, and the other end has a locking block 13. The locking block 13 engages within the slot 12 to achieve a snap-fit ​​connection between the first and second strap bodies. For example... Figure 5 As shown, the first belt has two ultrasonic sensors 11, the second belt has two ultrasonic sensors 11, and the four ultrasonic sensors 11 are evenly distributed along the circumference of the strap 10.

[0078] The strap 10 is loop-shaped and is used to bind to the target muscle group on the human body, ensuring that the ultrasonic sensor 11 fits tightly against the body. Along the circumference of the strap 10, the four ultrasonic sensors 11 are sequentially designated as the first, second, third, and fourth ultrasonic sensors. In use, the first ultrasonic sensor receives both the transmitted and echo signals; the second ultrasonic sensor performs the same operation, and so on. After the fourth ultrasonic sensor completes its corresponding operation, the cycle begins again. Because the four ultrasonic sensors are positioned differently on the target muscle group, this array acquisition and timing control method allows for comprehensive measurement of muscle deformation within the target muscle group.

[0079] like Figure 4 As shown, the exoskeleton robot includes a back module 31, a lower leg module 33, a thigh module 32, a support frame 20, and a foot module 34. The back module 31, thigh module 32, lower leg module 33, and foot module 34 are connected sequentially. The back module 31 is mounted on the support frame 20, and wheels 21 are provided at the bottom of the support frame 20 for movement. A hip joint torque sensor 41 is located at the connection between the back module 31 and the thigh module 32; a knee joint torque sensor 42 is located at the connection between the thigh module 32 and the lower leg module 33; and an ankle joint torque sensor 43 is located at the connection between the foot module 34 and the lower leg module 33. The hip joint encoder 44 is provided at the connection between the back module 31 and the thigh module 32, the knee joint encoder 45 is provided at the connection between the thigh module 32 and the calf module 33, and the ankle joint encoder 46 is provided at the connection between the foot module 34 and the calf module 33. The angle and position of each joint are known in real time by the hip joint encoder 44, the knee joint encoder 45 and the ankle joint encoder 46, so as to perform closed-loop control of the exoskeleton robot.

[0080] The exoskeleton robot system provided in this invention uses ultrasonic sensors to detect changes in deep muscle groups in the human body. Compared with traditional electromyography (EMG) sensors, it is more reliable and suitable for people with weak muscle strength. Traditional EMG sensors are cumbersome to calibrate, while ultrasonic sensors have high human-machine tolerance and high numerical consistency.

[0081] Optionally, the strap 10 can be detachably mounted on the bracket 20 or structurally separate from the exoskeleton robot.

[0082] Optionally, there are multiple straps 10, which are used to secure the body at different locations.

[0083] For example, there are two straps 10. In use, one strap 10 is tied above the knee joint and the other strap 10 is tied above the ankle joint. Of course, there can also be one, two, or four straps 10, and this embodiment of the invention does not specifically limit the number of straps 10.

[0084] In one specific embodiment of this application, the ultrasonic sensor 11 includes a host computer, a main control chip, a high-voltage pulse transmitting chip, and an echo receiving module. The main control chip is communicatively connected to the host computer and is used to control the operating mode and pulse duration of the high-voltage pulse transmitting chip. The echo receiving module is used to receive the ultrasonic echo signal and is communicatively connected to the main control chip.

[0085] The main control chip transmits a 3.3v control signal to control the working mode and pulse length of the HV7331 high-voltage pulse transmitting chip. A boost module is arranged between the high-voltage pulse chip and the main control chip. Specifically, the main control chip controls the LT8361 boost module to output a 70V high voltage, and the ultrasonic signal transmitting end in the high-voltage pulse transmitting chip is connected with the output end of the boost module to convert the high-voltage electrical signal into an ultrasonic pulse with a specific pulse width and frequency, so as to realize the transmission of the ultrasonic pulse. Optionally, the ultrasonic probe in the high-voltage pulse transmitting chip is multiplexed as an ultrasonic echo receiving module. Specifically, the echo receiving module is designed based on the AFE5805, and the original ultrasonic echo signal collected is filtered, amplified and analog-digital converted, and then sent to the upper computer by the main control chip. The upper computer calculates the deformation of the human muscle group based on the limit difference between the transmission signal and the return signal of the ultrasonic wave.

[0086] On the basis of the above embodiment, the ultrasonic sensor 11 further comprises a transceiver isolation module, which is arranged between the high-voltage pulse transmitting chip and the echo receiving module.

[0087] Specifically, the transceiver isolation module based on MD0105 is designed between the high-voltage pulse transmitting chip and the echo receiving module to prevent the high-voltage pulse signal of the transmitting end of the high-voltage pulse transmitting chip from affecting the echo receiving module, thereby protecting the echo receiving module.

[0088] Figure 6 The controller provided in the embodiment of the present application has the structure as shown in the structural schematic diagram of the controller, Figure 6 As shown in the structural schematic diagram of the controller, the controller can include a processor 610, a communications interface 620, a memory 630 and a communications bus 640, wherein the processor 610, the communications interface 620 and the memory 630 complete mutual communication through the communications bus 640. The processor 610 can invoke the logic command in the memory 630 to execute the following method: obtaining the deformation of the human muscle group; inputting the deformation into an intention recognition model to obtain the action intention of the human body output by the intention recognition model; inputting the deformation into a muscle force mapping model to obtain the muscle force of the muscle group output by the muscle force mapping model; and determining the control instruction of each joint of the exoskeleton robot based on the action intention and the muscle force of the muscle group. The intention recognition model is trained by taking the deformation sample data of the human muscle group as a sample and taking the action intention of the human body corresponding to the deformation as a label. The muscle force mapping model is trained by taking the deformation sample data of the human muscle group as a sample and taking the muscle force corresponding to the deformation sample data as a label.

[0089] In addition, the logic commands in the memory 630 described above can be implemented in the form of a software function unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of commands to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0090] The embodiment of the present application also provides a non-transitory computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method provided by the above-mentioned embodiments, for example, comprising: obtaining a deformation variable of a human muscle group; inputting the deformation variable into an intention recognition model to obtain an action intention of the human body output by the intention recognition model; inputting the deformation variable into a muscle force mapping model to obtain muscle force of the muscle group output by the muscle force mapping model; and determining a control instruction of each joint of an exoskeleton robot based on the action intention and the muscle force of the muscle group. The intention recognition model is trained by taking the deformation variable sample data of the human muscle group as a sample and taking the action intention of the human body corresponding to the deformation variable as a label. The muscle force mapping model is trained by taking the deformation variable sample data of the human muscle group as a sample and taking the muscle force corresponding to the deformation variable sample data as a label.

[0091] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A control method for an exoskeleton robot, characterized in that, include: Obtain the deformation of human muscle groups; The deformation is input into the intention recognition model to obtain the human body's action intention output by the intention recognition model; The deformation variables are input into the muscle strength mapping model to obtain the muscle group strength output by the muscle strength mapping model. Based on the stated intention of movement and the muscle strength of the muscle groups, control commands for each joint of the exoskeleton robot are determined. The intention recognition model is trained based on the deformation sample data of human muscle groups and the action intention sample data of human beings; the muscle strength mapping model is trained based on the deformation sample data of human muscle groups and the muscle strength sample data of muscle groups. The specific control commands for each joint of the exoskeleton robot determined based on the action intention and the muscle group strength include: The resultant torque of each joint under human-machine coupling is determined based on the stated action intention and walking dynamics model; the walking dynamics model characterizes the correspondence between the standard gait of the human body and the torque of each joint of the exoskeleton robot under human-machine coupling. The output torque of each joint in the human body is determined based on the muscle strength of the aforementioned muscle groups. The compensation torque of each joint of the exoskeleton robot is determined based on the resultant torque and the output torque, and the movement of each joint of the exoskeleton robot is controlled based on the compensation torque.

2. The exoskeleton robot control method according to claim 1, characterized in that, Before obtaining the deformation of human muscle groups, the following steps are also included: Acquire the transmitted and transmitted signals from multiple ultrasonic sensors; The deformation of human muscle groups is determined based on the phase difference between the transmitted signal and the transmitted signal. The ultrasonic sensors are positioned at different locations on the human body.

3. The exoskeleton robot control method according to claim 1, characterized in that, The specific steps of inputting the deformation variable into the intent recognition model to obtain the action intent output by the intent recognition model include: The deformation is input into the action recognition model to obtain the human action output by the action recognition model; The human body action is input into the intention mapping model to obtain the action intention output by the intention mapping model; The action recognition model is trained based on human muscle group deformation sample data and human action sample data; the intention mapping model is trained based on human action sample data and action intention sample data.

4. An exoskeleton robot system applied to the exoskeleton robot control method according to any one of claims 1 to 3, characterized in that, The invention includes an exoskeleton robot and a strap. Each joint of the exoskeleton robot is equipped with a torque sensor, the strap is equipped with multiple ultrasonic sensors, and a controller is provided on the exoskeleton robot. The ultrasonic sensors and the torque sensors are all communicatively connected to the controller.

5. The exoskeleton robot system according to claim 4, characterized in that, The multiple ultrasonic sensors are evenly distributed along the circumference of the strap.

6. The exoskeleton robot system according to claim 4, characterized in that, The ultrasonic sensor includes a host computer, a main control chip, a high-voltage pulse transmitting chip, and an echo receiving module. The main control chip is communicatively connected to the host computer and is used to control the working mode and pulse duration of the high-voltage pulse transmitting chip. The echo receiving module is used to receive the echo signal of the ultrasonic wave and is communicatively connected to the main control chip.

7. The exoskeleton robot system according to claim 6, characterized in that, The ultrasonic sensor also includes a transceiver isolation module, which is disposed between the high-voltage pulse transmitting chip and the echo receiving module.

8. A controller, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the exoskeleton robot control method as described in any one of claims 1 to 3.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the exoskeleton robot control method as described in any one of claims 1 to 3.

Images