Lower limb rehabilitation robot knee joint driving control system based on machine vision

Abstract

The invention belongs to the technical field of machine vision control, and particularly relates to a lower limb rehabilitation robot knee joint driving control system based on machine vision, which comprises a central processing module, a machine vision module, a driving control module, a feedback and adjustment module and a man-machine interaction module. A training target can be set, the progress can be checked, parameters can be adjusted, closed-loop control over the whole system can be achieved, motion state monitoring is provided through machine vision, the control system adjusts the driving device according to feedback, and the safety in the training process is guaranteed by monitoring the motion condition of knee joints in real time and adjusting the driving force according to the feedback of a patient.

Description

Technical Field

[0001] The present invention relates to the field of machine vision control and processing, and particularly to a knee joint drive control system of a lower limb rehabilitation robot based on machine vision. Background Art

[0002] Lower limb rehabilitation robot technology is a newly emerging technology that has developed rapidly in recent years. It is a new application of robot technology in the medical field and has gradually developed into a new motor nerve rehabilitation treatment technology. Standing, walking, and going up and down stairs are movements that are repeated frequently in daily life. However, for patients with lower limb paralysis, muscle damage, spinal cord damage, and elderly patients with insufficient lower limb muscle strength, they cannot complete these movements. In recent years, with the increasing number of patients with lower limb movement disorders year by year and the trend of aging, the development of the rehabilitation industry has become very urgent. And the quality of life of patients depends on the degree of limb function recovery. How to use modern advanced rehabilitation treatment technologies to improve the limb movement function of patients and enable patients to regain their ability to live independently as soon as possible while getting rid of the torture of disability has always been the focus of research and practice by rehabilitation workers;

[0003] As the name implies, a lower limb rehabilitation treatment robot is a mechanical device that facilitates the human body to cyclically lift and lower the lower limbs for human body rehabilitation treatment. However, in the existing technologies, they are all manually operated or controlled by a single motor to perform cyclic operations of lifting and lowering. However, its manual operation makes it quite difficult and inconvenient for the rehabilitation personnel with inconvenient legs and feet to operate by themselves. Therefore, there is an urgent need to provide a knee joint drive control system of a lower limb rehabilitation robot based on machine vision. Summary of the Invention

[0004] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Simplifications or omissions may be made in this part, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this part, the abstract, and the title. However, such simplifications or omissions shall not be used to limit the scope of the present invention.

[0005] Therefore, the purpose of the present invention is to provide a knee joint drive control system of a lower limb rehabilitation robot based on machine vision. The entire system realizes closed-loop control. Machine vision provides motion state monitoring. The control system adjusts the drive device according to the feedback, monitors the motion of the knee joint in real time, and adjusts the driving force according to the feedback of the patient to ensure the safety during the training process.

[0006] To solve the above technical problems, according to one aspect of the present invention, the following technical solutions are provided:

[0007] A knee joint drive control system of a lower limb rehabilitation robot based on machine vision, comprising:

[0008] Central processing module: As the instruction control end of the system, it ensures that the machine executes programs in a specified order, generates operation signals for each instruction fetched from the memory, sends various operation signals to corresponding components, and controls each component to act according to the requirements of the instructions;

[0009] Machine vision module: Used to capture the position, angle, and movement trajectory of the patient's knee joint in real time;

[0010] Motion analysis and pose estimation module: Processes the images collected by the vision module to analyze the motion state of the patient's knee joint;

[0011] Drive control module: Precisely controls the drive of the knee joint according to the motion analysis results;

[0012] Feedback and adjustment module: Provides real-time motion feedback, adjusts parameters during the rehabilitation process, and ensures the safety and effectiveness of training;

[0013] Human-machine interaction module: Provides a friendly interaction interface for patients and operators, used to set training goals, view progress, and adjust parameters.

[0014] As a preferred solution of a knee joint drive control system for a lower limb rehabilitation robot based on machine vision according to the present invention, wherein: the machine vision module is used to acquire motion images of the patient's knee joint, perform image processing and analysis, and consists of the following sub-modules:

[0015] Camera: A high-resolution camera, such as an RGB camera, a depth camera, or an infrared sensor, is installed at a fixed position to capture dynamic information of the patient's knee joint. Multiple cameras capture knee joint movements from different angles to ensure comprehensive motion data acquisition;

[0016] Image processing unit: Uses image processing algorithms, such as edge detection, image segmentation, and feature extraction, to process the acquired images and extract information on the position, angle, and movement trajectory of the knee joint;

[0017] Depth sensor: If the position of the knee joint in three-dimensional space needs to be acquired, depth sensors, such as lidar, structured light, and ToF sensors, can provide accurate depth information to help estimate the spatial position of the knee joint.

[0018] As a preferred solution of a knee joint drive control system for a lower limb rehabilitation robot based on machine vision according to the present invention, wherein: the motion analysis and pose estimation module is used to analyze the image data provided by the machine vision module to identify the motion state of the patient's knee joint and estimate its angle and bending degree parameters. The motion analysis and pose estimation include the following functions:

[0019] Joint angle calculation: By performing feature matching and motion tracking on the images of the knee joint, the bending angle of the knee joint, such as the flexion and extension angle of the knee joint, is calculated in real time to understand the patient's range of motion;

[0020] Motion trajectory tracking: Analyze the motion trajectory of the knee joint to determine whether the patient follows the predetermined path or plan during the rehabilitation training;

[0021] Posture recognition: Estimate the posture of the patient's lower limbs based on visual information, such as knee angle, sitting or standing posture, to judge whether the motion is standard.

[0022] As a preferred solution of a knee joint drive control system of a lower limb rehabilitation robot based on machine vision according to the present invention, wherein: the drive control module is used to execute by receiving signals from the motion analysis module, control the drive of the patient's knee joint by the lower limb rehabilitation robot, so as to achieve the rehabilitation training goal, and the module generally includes:

[0023] Motor and servo driver: The drive system uses a servo motor or a stepper motor to drive the motion of the knee joint, and adjusts the speed, torque and direction of the motor according to the calculation results of the motion analysis module;

[0024] Control algorithm: The control system adjusts the output of the driver according to the actual motion requirements of the patient. Common control algorithms include PID control, fuzzy control, and motion planning control, and adjust the driving force of the knee joint according to the patient's real-time motion state;

[0025] Torque and force sensor: To ensure the safety and comfort of the motion, the system uses a torque sensor to detect the force exerted by the knee joint driver to avoid causing harm to the patient due to excessive force.

[0026] As a preferred solution of a knee joint drive control system of a lower limb rehabilitation robot based on machine vision according to the present invention, wherein: the feedback and adjustment module is used to execute real-time feedback and adjustment functions to ensure the personalization, accuracy and safety of the patient's motion training, and the functions include:

[0027] Motion progress feedback: Provide real-time feedback on the current motion state of the patient's knee joint, including bending angle and motion speed parameters, and provide them to the patient through a display screen or voice prompt;

[0028] Personalized adjustment: Dynamically adjust the training intensity, training time, and knee joint angle range according to the patient's rehabilitation needs and progress, and cooperate with the operator or patient through the human-computer interaction module for adjustment;

[0029] Safety warning: When an abnormal situation is detected, such as excessive motion or excessive force, the system issues a warning and adjusts the output of the driver in a timely manner to ensure the safety of the patient.

[0030] As a preferred solution of the knee joint drive control system of the lower limb rehabilitation robot based on machine vision according to the present invention, wherein: the human-computer interaction module is used to provide a friendly interface for the patient, for inputting training goals, viewing real-time progress, and adjusting training parameters. The module usually includes:

[0031] Touch screen interface: The user can input personal information and set training goals, such as knee joint bending angle and training time, through the touch screen, and view the movement progress and feedback;

[0032] Voice assistant: The system is configured with a voice assistant to help the patient operate and control more conveniently;

[0033] Training plan management: According to the patient's rehabilitation plan, provide various training modes or personalized rehabilitation programs, such as knee flexion training, knee extension training, and flexibility training.

[0034] As a preferred solution of the knee joint drive control system of the lower limb rehabilitation robot based on machine vision according to the present invention, wherein: the control method of the knee joint drive control system includes the following steps:

[0035] S1 Image acquisition and processing: Set multiple cameras near the patient's lower limb to obtain the motion images of the knee joint in real time, and perform preprocessing on the collected original images, such as denoising, light compensation, and contrast enhancement, for subsequent feature extraction. At the same time, identify the key feature points of the knee joint through image processing algorithms, and use the pose estimation method in computer vision to track the position and angle of the knee joint in real time;

[0036] S2. Motion analysis and pose estimation: Based on the results of image recognition and feature extraction, calculate the bending angle or extension angle of the knee joint, track the continuous movement of the knee joint, analyze the motion trajectory of the knee joint, including movements such as bending and extension, and identify whether the knee joint is undergoing rehabilitation training, and judge whether the current action conforms to the predetermined rehabilitation plan;

[0037] S3. Control strategy and algorithm: Set the training goals of the knee joint according to the patient's rehabilitation needs and current motor ability, and dynamically adjust the training goals according to the patient's rehabilitation progress to ensure that the training content is within the scope of their ability and avoid overtraining or inappropriate exercise intensity;

[0038] S4. Control algorithm design: Based on the deviation between the knee joint angle and the desired target, adopt the PID control algorithm to accurately control the output of the knee joint driver, so that its movement follows the set target angle change, and adopt the trajectory planning algorithm to plan the movement path of the knee joint to ensure that the knee joint can flex and extend smoothly and naturally during the rehabilitation training process;

[0039] S5. Drive control: According to the output of the control algorithm, drive the servo motor in the system to perform corresponding movements to drive the flexion and extension of the knee joint, enabling the servo motor to adjust the torque and angle according to requirements to ensure the precise movement of the knee joint. Moreover, monitor the force exerted on the knee joint by the driver through a torque sensor to ensure that the force will not be too large to avoid harm to the patient. If the system detects excessive force, the driver will immediately slow down or stop the action.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0041] 1. The entire system realizes closed-loop control. Machine vision provides motion state monitoring, and the control system adjusts the drive device according to the feedback. By monitoring the motion of the knee joint in real time and adjusting the driving force according to the patient's feedback, the safety during the training process is ensured.

[0042] 2. The machine vision and motion analysis module can accurately obtain the motion state of the knee joint and provide more refined drive control.

[0043] 3. It can dynamically adjust the training intensity and method according to the specific conditions of the patient, such as the motion range of the knee joint and the rehabilitation progress, and customize the rehabilitation plan personalized. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the present invention will be described in detail below in conjunction with the drawings and specific embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them:

[0045] Figure 1 It is the block diagram of the knee joint drive control system of the present invention;

[0046] Figure 2 It is the schematic diagram of the control steps of the control system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following detailed description of the specific embodiments of the present invention will be made in conjunction with the drawings.

[0048] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar promotions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0049] Secondly, the present invention will be described in detail with reference to the schematic diagrams. When describing the embodiments of the present invention in detail, for the convenience of explanation, the cross-sectional views showing the device structure will be locally enlarged out of the general scale, and the schematic diagrams are only examples and should not limit the scope of protection of the present invention herein. In addition, in actual production, three-dimensional spatial dimensions including length, width, and depth should be included.

[0050] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0051] The present invention provides a knee joint drive control system for a lower limb rehabilitation robot based on machine vision. Please refer to Figure 1-2 , including a knee joint drive control system for a lower limb rehabilitation robot based on machine vision, which is characterized by including:

[0052] Central processing module: As the instruction control end of the system, it ensures that the machine executes the program in a specified order, generates the operation signals for each instruction fetched from the memory, sends various operation signals to the corresponding components, and controls each component to act according to the requirements of the instruction;

[0053] Machine vision module: Used to capture the position, angle, and movement trajectory of the patient's knee joint in real time;

[0054] Motion analysis and pose estimation module: Processes the images collected by the vision module to analyze the motion state of the patient's knee joint;

[0055] Drive control module: Precisely controls the drive of the knee joint according to the motion analysis results;

[0056] Feedback and regulation module: Provides real-time motion feedback, adjusts the parameters during the rehabilitation process, and ensures the safety and effectiveness of the training;

[0057] Human-machine interaction module: Provides a friendly interaction interface for patients and operators, and is used to set training goals, view progress, and adjust parameters;

[0058] The machine vision module is used to obtain the motion images of the patient's knee joint, and perform image processing and analysis, and is composed of the following sub-modules:

[0059] Camera: A high-resolution camera, such as an RGB camera, a depth camera, or an infrared sensor, is installed at a fixed position to capture the dynamic information of the patient's knee joint. Multiple cameras capture the knee joint movement from different angles to ensure comprehensive motion data acquisition;

[0060] Image processing unit: Uses image processing algorithms, such as edge detection, image segmentation, and feature extraction, to process the acquired images, and extracts the position, angle, and movement trajectory information of the knee joint;

[0061] Depth sensor: If it is necessary to obtain the position of the knee joint in three-dimensional space, depth sensors such as lidar, structured light, and ToF sensors can provide accurate depth information to help estimate the spatial position of the knee joint;

[0062] The motion analysis and pose estimation module is used to analyze the image data provided by the machine vision module to identify the motion state of the patient's knee joint and estimate its angle and flexion parameters. The motion analysis and pose estimation include the following functions:

[0063] Joint angle calculation: By performing feature matching and motion tracking on the images of the knee joint, the flexion angle of the knee joint, such as the flexion and extension angle of the knee joint, is calculated in real time to understand the patient's range of motion;

[0064] Motion trajectory tracking: Analyze the motion trajectory of the knee joint to determine whether the patient follows the predetermined path or plan during the rehabilitation training;

[0065] Pose recognition: Estimate the pose of the patient's lower limb based on visual information, such as knee angle, sitting or standing posture, to determine whether the motion is standard;

[0066] The drive control module is used to control the drive of the lower limb rehabilitation robot for the patient's knee joint by receiving signals from the motion analysis module, so as to achieve the rehabilitation training goal. The module usually includes:

[0067] Motor and servo driver: The drive system uses a servo motor or a stepper motor to drive the motion of the knee joint, and adjusts the speed, torque, and direction of the motor according to the calculation results of the motion analysis module;

[0068] Control algorithm: The control system adjusts the output of the driver according to the actual motion requirements of the patient. Common control algorithms include PID control, fuzzy control, and motion planning control, and adjust the driving force of the knee joint according to the patient's real-time motion state;

[0069] Torque and force sensor: To ensure the safety and comfort of the motion, the system uses a torque sensor to detect the force exerted by the knee joint driver to avoid causing harm to the patient due to excessive force;

[0070] The feedback and regulation module is used to perform real-time feedback and regulation functions to ensure the personalization, accuracy, and safety of the patient's motion training. The functions include:

[0071] Motion progress feedback: Provide real-time feedback on the current motion state of the patient's knee joint, including flexion angle and motion speed parameters, and provide them to the patient through a display screen or voice prompt;

[0072] Personalized adjustment: Dynamically adjust the training intensity, training time, and knee joint angle range according to the patient's rehabilitation needs and progress, and cooperate with the operator or patient to make adjustments through the human-machine interaction module;

[0073] Safety warning: When abnormal situations are detected, such as excessive exercise or excessive force, the system issues a warning and adjusts the output of the driver in a timely manner to ensure the safety of the patient;

[0074] The human-machine interaction module is used to provide a friendly interface for the patient to input training goals, view real-time progress, and adjust training parameters. The module usually includes:

[0075] Touch screen interface: Users can input personal information and set training goals, such as knee joint bending angle and training time, through the touch screen, and view the exercise progress and feedback;

[0076] Voice assistant: The system is configured with a voice assistant to help the patient operate and control more conveniently;

[0077] Training plan management: Provide multiple training modes or personalized rehabilitation programs according to the patient's rehabilitation plan, such as knee flexion training, knee extension training, and flexibility training;

[0078] The control method of the knee joint drive control system includes the following steps:

[0079] S1 Image acquisition and processing: Set multiple cameras near the patient's lower limbs to obtain real-time motion images of the knee joint, and perform preprocessing on the collected original images, such as denoising, light compensation, and contrast enhancement, for subsequent feature extraction. At the same time, identify the key feature points of the knee joint through image processing algorithms, and use the pose estimation method in computer vision to track the position and angle of the knee joint in real time;

[0080] S2. Motion analysis and pose estimation: Based on the results of image recognition and feature extraction, calculate the bending angle or extension angle of the knee joint, track the continuous motion of the knee joint, analyze the motion trajectory of the knee joint, including motions such as bending and extension, and identify whether the knee joint is undergoing rehabilitation training, and judge whether the current action conforms to the predetermined rehabilitation plan;

[0081] S3. Control strategy and algorithm: Set the training goals of the knee joint according to the patient's rehabilitation needs and current motor ability, and dynamically adjust the training goals according to the patient's rehabilitation progress to ensure that the training content is within the scope of their ability and avoid overtraining or inappropriate exercise intensity;

[0082] S4. Control algorithm design: Based on the deviation between the knee joint angle and the desired target, the PID control algorithm is adopted to accurately control the output of the knee joint driver, enabling its movement to follow the set target angle change. And the trajectory planning algorithm is used to plan the movement path of the knee joint to ensure that the knee joint can flex and extend smoothly and naturally during the rehabilitation training.

[0083] S5. Drive control: According to the output of the control algorithm, the servo motor in the drive system performs corresponding movements to drive the flexion and extension of the knee joint, enabling the servo motor to adjust the torque and angle according to the demand to ensure the accurate movement of the knee joint. And the force exerted on the knee joint by the driver is monitored through the torque sensor to ensure that the force will not be too large to avoid harming the patient. If the system detects excessive force, the driver will immediately slow down or stop operating.

[0084] Although the present invention has been described above with reference to the embodiments, various improvements can be made to it and components can be replaced with effective substances without departing from the scope of the present invention. In particular, as long as there is no structural conflict, the features in the embodiments disclosed in the present invention can be combined with each other in any way. The reason for not exhaustively describing the situations of these combinations in this specification is only to save space and resources. Therefore, the present invention is not limited to the specific embodiments disclosed in the text, but includes all technical solutions falling within the scope of the claims.

Claims

1. A lower limb rehabilitation robot knee joint drive control system based on machine vision, characterized in that: include: Central processing module: As the command control end of the system, it ensures that the machine executes the program in the prescribed order, generates operation signals for each instruction taken out from the memory, sends various operation signals to the corresponding components, and controls each component to act according to the requirements of the instructions; Machine vision module: used to capture the position, angle and motion trajectory of the patient's knee joint in real time; Motion analysis and posture estimation module: processes the images collected by the visual module and analyzes the motion state of the patient's knee joint; Drive control module: accurately controls the drive of the knee joint according to the results of motion analysis; Feedback and adjustment module: provides real-time movement feedback, adjusts parameters during rehabilitation, and ensures the safety and effectiveness of training; Human-computer interaction module: provides patients and operators with a friendly interactive interface for setting training goals, checking progress, and adjusting parameters.

2. The machine vision-based lower limb rehabilitation robot knee joint drive control system according to claim 1, characterized in that: The machine vision module is used to obtain the motion image of the patient's knee joint and perform image processing and analysis, and is composed of the following submodules: Camera: Use a high-resolution camera, such as an RGB camera, a depth camera, or an infrared sensor installed in a fixed position to capture dynamic information about the patient's knee joint. Multiple cameras capture knee joint motion from different angles to ensure comprehensive motion data. Image processing unit: uses image processing algorithms such as edge detection, image segmentation and feature extraction to process the acquired images and extract the position, angle and motion trajectory information of the knee joint; Depth sensor: If you need to obtain the position of the knee joint in three-dimensional space, depth sensors, such as lidar, structured light, and ToF sensors, can provide accurate depth information to help estimate the spatial position of the knee joint.

3. The machine vision-based lower limb rehabilitation robot knee joint drive control system according to claim 2, characterized in that: The motion analysis and posture estimation module is used to analyze the image data provided by the machine vision module to identify the motion state of the patient's knee joint and estimate its angle and curvature parameters. The motion analysis and posture estimation includes the following functions: Joint angle calculation: By performing feature matching and motion tracking on the knee joint image, the bending angle of the knee joint, such as the knee flexion and extension angle, is calculated in real time to understand the patient's range of motion; Movement trajectory tracking: Analyze the movement trajectory of the knee joint to determine whether the patient is performing rehabilitation training according to the predetermined path or plan during exercise; Posture recognition: Estimate the patient's lower limb posture based on visual information, such as knee angle, sitting or standing posture, to determine whether the movement is normal.

4. The machine vision-based lower limb rehabilitation robot knee joint drive control system according to claim 3, characterized in that: The drive control module is used to control the lower limb rehabilitation robot to drive the patient's knee joint by receiving the signal from the motion analysis module, so as to achieve the rehabilitation training goal. The module generally includes: Motor and servo drive: The drive system uses a servo motor or a stepper motor to drive the movement of the knee joint, and adjusts the speed, torque and direction of the motor according to the calculation results of the motion analysis module; Control algorithm: The control system adjusts the output of the driver according to the patient's actual movement needs. Common control algorithms include PID control, fuzzy control, and motion planning control. The driving force of the knee joint is adjusted according to the patient's real-time movement status. Torque and force sensors: To ensure safe and comfortable movement, the system uses torque sensors to detect the force applied by the knee joint actuator to prevent excessive force from causing harm to the patient.

5. The machine vision-based lower limb rehabilitation robot knee joint drive control system according to claim 4, characterized in that: The feedback and adjustment module is used to perform real-time feedback and adjustment functions to ensure the personalization, accuracy and safety of patient exercise training. The functions include: Movement progress feedback: Provides real-time feedback on the patient's current knee joint movement status, including bending angle and movement speed parameters, and provides it to the patient through a display screen or voice prompts; Personalized adjustment: Dynamically adjust the training intensity, training time, and knee joint angle range according to the patient's rehabilitation needs and progress, and coordinate with the operator or patient through the human-computer interaction module; Safety warning: When abnormal conditions are detected, such as excessive movement or force, the system issues a warning and adjusts the output of the driver in time to ensure the safety of the patient.

6. The machine vision-based lower limb rehabilitation robot knee joint drive control system according to claim 5, characterized in that: The human-computer interaction module is used to provide a friendly interface for patients to input training goals, view real-time progress, and adjust training parameters. The module generally includes: Touchscreen interface: Users can use the touchscreen to input personal information, set training goals, such as knee flexion angle, training time, and view exercise progress and feedback; Voice assistant: The system is equipped with a voice assistant to help patients operate and control more conveniently; Training plan management: According to the patient's rehabilitation plan, provide a variety of training modes or personalized rehabilitation plans, such as knee flexion training, knee extension training, and flexibility training.

7. The machine vision-based lower limb rehabilitation robot knee joint drive control system according to claim 6, characterized in that: The control method of the knee joint drive control system comprises the following steps: S1 Image acquisition and processing: multiple cameras are set up near the patient's lower limbs to obtain real-time motion images of the knee joint, and the collected original images are pre-processed by denoising, lighting compensation, contrast enhancement, etc. for subsequent feature extraction. At the same time, the key feature points of the knee joint are identified through image processing algorithms, and the position and angle of the knee joint are tracked in real time using posture estimation methods in computer vision; S2, motion analysis and posture estimation, based on the results of image recognition and feature extraction, calculate the bending angle or extension angle of the knee joint, track the continuous movement of the knee joint, analyze the movement trajectory of the knee joint, including bending, extension and other movements, and identify whether the knee joint is undergoing rehabilitation training, and judge whether the current movement conforms to the predetermined rehabilitation plan; S3, control strategy and algorithm, set knee joint training goals according to the patient's rehabilitation needs and current exercise ability, and dynamically adjust the training goals according to the patient's rehabilitation progress to ensure that the training content is in line with their ability range and avoid overtraining or inappropriate exercise intensity; S4. Control algorithm design: Based on the deviation between the knee joint angle and the desired target, a PID control algorithm is used to accurately control the output of the knee joint driver so that its movement follows the set target angle. A trajectory planning algorithm is used to plan the knee joint movement path to ensure that the knee joint can be flexed and extended smoothly and naturally during rehabilitation training. S5, drive control, according to the output of the control algorithm, the servo motor in the drive system performs corresponding movements to promote the flexion and extension of the knee joint, so that the servo motor adjusts the torque and angle according to the needs to ensure the precise movement of the knee joint, and monitors the force applied by the driver on the knee joint through the torque sensor to ensure that the force is not too large to avoid harm to the patient. If the system detects excessive force, the driver will immediately slow down or stop the action.

Images