Lower limb rehabilitation exoskeleton robot control method and system

Abstract

The application provides a lower limb rehabilitation exoskeleton robot control method and system, which comprises the following steps: collecting joint torque, lower limb joint angle, joint angular velocity and plantar pressure data of a patient to calculate a rehabilitation state evaluation index; collecting the angle difference of left and right hip joints, identifying the current gait type and the expected trajectory under the corresponding gait based on the gait phase variable, and optimizing the expected trajectory based on the rehabilitation state evaluation index to obtain a target optimized trajectory; calculating the nominal torque based on the acceleration of the trajectory and the parameters of the dynamics model, designing a sliding surface through the implementation of monitoring trajectory tracking error, generating a compensation torque, and calculating a control torque vector. By identifying the gait type and generating a personalized trajectory, and then decomposing the nominal torque and the compensation torque through the U-K controller, high-precision tracking is realized while the model uncertainty disturbance is inhibited, and the naturalness of human-machine cooperation and the safety of rehabilitation training are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of exoskeleton robot technology, and in particular to a control method and system for a lower limb rehabilitation exoskeleton robot. Background Technology

[0002] Lower limb rehabilitation exoskeleton robots are primarily designed for patients with lower limb motor dysfunction caused by neurological diseases such as stroke and spinal cord injury. They aim to rebuild neuromuscular pathways, improve patients' walking ability, and prevent secondary injuries through task-oriented repetitive gait training.

[0003] Existing conventional lower limb rehabilitation exoskeleton robot control methods have the following significant limitations: On the one hand, training mode switching relies on manual adjustment by the therapist and lacks autonomous decision-making ability based on the patient's real-time status. For example, intention recognition based on electromyographic signals is easily affected by sweating and electrode displacement, leading to misjudgment of intention. On the other hand, control strategies mostly adopt fixed trajectory tracking or rigid impedance control, which is difficult to adapt to individual gait differences, easily causing human-machine movement incoordination, producing a strong pulling sensation or even secondary joint damage, and affecting rehabilitation efficiency. Summary of the Invention

[0004] In view of the shortcomings of the prior art, the purpose of this invention is to provide a control method and system for a lower limb rehabilitation exoskeleton robot, which aims to solve the technical problem of low rehabilitation efficiency in the existing control methods.

[0005] To achieve the above objectives, in a first aspect, the present invention provides: a control method for a lower limb rehabilitation exoskeleton robot, comprising the following steps:

[0006] The patient's joint torque, lower limb joint angle, joint angular velocity, and plantar pressure data are collected through a joint encoder, plantar pressure sensor, and inertial measurement unit.

[0007] Rehabilitation status evaluation indicators are calculated based on joint torque, lower limb joint angle, and plantar pressure data.

[0008] The angle difference between the left and right hip joints is collected to construct gait phase variables that characterize the gait cycle. The time series data composed of joint angular velocity and plantar pressure data is input into a pre-trained temporal classification model to identify the current gait type.

[0009] The desired trajectory under the corresponding gait is matched according to the gait type, and the desired trajectory is optimized based on the rehabilitation status evaluation index to obtain the target optimized trajectory;

[0010] A dynamic model of the exoskeleton coupled with the human body is established, and the target optimized trajectory is used as the servo constraint target. An acceleration constraint equation is constructed, and the nominal torque is calculated based on the dynamic model parameters and the acceleration of the trajectory. By implementing the monitoring of trajectory tracking error, a sliding surface is designed to generate a compensation torque, so as to calculate the control torque vector.

[0011] According to one aspect of the above technical solution, the calculation expression for the rehabilitation status evaluation index is as follows:

[0012] ;

[0013] In the formula, V is the rehabilitation status evaluation index, and M is the joint torque. This is a torque reference value. R is the first weighting coefficient for joint torque, and R is the lower limb joint angle. For angle reference value, The second weighting coefficient for lower limb joint angles is denoted by S, where S represents plantar pressure data. This is a pressure reference value. This is the third weighting coefficient for plantar pressure data.

[0014] According to one aspect of the above technical solution, the second weighting coefficient and the third weighting coefficient are dynamically adjusted based on gait phase variables, and the calculation expression for the adjusted weighting coefficients is as follows:

[0015] ;

[0016] ;

[0017] In the formula, This is the adjusted second weighting coefficient. This serves as the baseline value for the second weighting coefficient. This is the adjusted third weighting coefficient. This is the baseline value for the third weighting coefficient. For gait phase variables.

[0018] According to one aspect of the above technical solution, the steps for constructing gait phase variables characterizing gait cycles specifically include:

[0019] The angle difference between the left and right hip joints is low-pass filtered, and the filtered angle difference is normalized to an interval. Then, the phase variable is calculated by integration. The expression for calculating the gait phase variable is as follows:

[0020] ;

[0021] In the formula, For gait phase variables, For gait period, This is the angle value of the left hip joint. t represents the angle of the right hip joint;

[0022] The update expression for the gait period is:

[0023] ;

[0024] In the formula, For the updated gait cycle, Forgetting factor, For the previous state period, This represents the current gait cycle.

[0025] According to one aspect of the above technical solution, the calculation expression of the dynamic model is:

[0026] ;

[0027] In the formula, , , These represent the angle, angular velocity, and acceleration vectors of the joint, respectively. Represents the inertia matrix. Represents the matrix of Coriolis force and centripetal force. Represents the gravity term. Represents the control torque vector;

[0028] The expression for the constraint equation of acceleration is:

[0029] ;

[0030] In the formula, The constraint matrix represents the geometric relationship associated with the desired trajectory. This represents the constraint vector related to the velocity and acceleration of the trajectory.

[0031] According to one aspect of the above technical solution, the calculation expression for the control torque vector is:

[0032] ;

[0033] ;

[0034] ;

[0035] ;

[0036] ;

[0037] In the formula, This represents the nominal torque. Indicates expected acceleration, Indicates the desired speed. This indicates the compensation torque. Indicates sliding mode gain. Indicates the sliding surface. Represents the sliding mode control function. Indicates the reciprocal of the error. Indicates the parameters of the sliding surface. Indicates tracking error. Indicates the initial sliding mode gain. Indicates the learning rate. This represents the sliding surface norm.

[0038] According to one aspect of the above technical solution, after obtaining the control torque vector, the method further includes:

[0039] The sensor data of joint angular velocity and angular acceleration are acquired in real time, the gait phase angle is calculated by the positive and negative tangent functions, and the results are defuzzified to the gait interval to match the gait cycle.

[0040] Based on the gait angle, the assist torque is calculated using a Gaussian kernel function. The assist amplitude coefficient is dynamically linked to the rehabilitation status evaluation index. When the rehabilitation status evaluation index is greater than the preset value, the assist amplitude coefficient that determines the maximum peak value of the assist torque is linearly reduced to zero.

[0041] According to one aspect of the above technical solution, the calculation expression for the assist torque is:

[0042] ;

[0043] ;

[0044] In the formula, To assist torque, The assist amplitude coefficient, For gait phase angle, The preset gait phase angle, To control the width parameter of the assist torque distribution around a preset gait phase angle, The joint angular velocity, This is the joint angular acceleration.

[0045] Secondly, this solution also provides a lower limb rehabilitation exoskeleton robot control system, including:

[0046] The data module is used to collect the patient's joint torque, lower limb joint angle, joint angular velocity and plantar pressure data through joint encoders, plantar pressure sensors and inertial measurement units;

[0047] The evaluation module is used to calculate rehabilitation status evaluation indicators based on joint torque, lower limb joint angle and plantar pressure data;

[0048] The gait module is used to collect the angle difference between the left and right hip joints, construct gait phase variables that characterize the gait cycle, and input the time series of joint angular velocity and plantar pressure data into the pre-trained temporal classification model to identify the current gait type.

[0049] The trajectory module is used to match the expected trajectory under the corresponding gait type according to the gait type, and to optimize the expected trajectory based on the rehabilitation status evaluation index to obtain the target optimized trajectory;

[0050] The control module is used to establish a dynamic model of the exoskeleton coupled with the human body, and to construct an acceleration constraint equation with the target optimized trajectory as the servo constraint target. Based on the dynamic model parameters and the acceleration of the trajectory, the nominal torque is calculated, and a sliding surface is designed by monitoring the trajectory tracking error to generate a compensation torque in order to calculate the control torque vector.

[0051] According to one aspect of the above technical solution, the gait module is specifically used for:

[0052] The angle difference between the left and right hip joints is low-pass filtered, and the filtered angle difference is normalized to an interval. Then, the phase variable is calculated by integration. The expression for calculating the gait phase variable is as follows:

[0053] ;

[0054] In the formula, For gait phase variables, For gait period, This is the angle value of the left hip joint. t represents the angle of the right hip joint;

[0055] The update expression for the gait period is:

[0056] ;

[0057] In the formula, For the updated gait cycle, Forgetting factor, For the previous state period, This represents the current gait cycle.

[0058] According to one aspect of the above technical solution, the system further includes:

[0059] The adjustment module is used to acquire sensor data of joint angular velocity and angular acceleration in real time, calculate gait phase angle through tangent and inverse tangent functions, and deblur the results to the gait interval to match the gait cycle;

[0060] Based on the gait angle, the assist torque is calculated using a Gaussian kernel function. The assist amplitude coefficient is dynamically linked to the rehabilitation status evaluation index. When the rehabilitation status evaluation index is greater than the preset value, the assist amplitude coefficient that determines the maximum peak value of the assist torque is linearly reduced to zero.

[0061] Compared with existing technologies, the beneficial effects of this invention are as follows: by synchronously collecting joint motion parameters, plantar pressure, and inertial data, a linkage relationship between rehabilitation status indicators and gait phase is constructed, accurately identifying gait types and generating personalized trajectories. Then, the nominal torque and robust compensation torque are decomposed through the UK (Udwadia-Kalaba) controller, achieving high-precision tracking while suppressing model uncertainty disturbances. Finally, the amplitude and timing of the assist torque are adaptively adjusted in conjunction with gait phase, thereby forming an "assessment-planning-compensation" closed loop. This effectively solves the problems of poor robustness of intent recognition, lack of flexibility in control rigidity, and disconnect between rehabilitation strategies and physiological timing in existing technologies, significantly improving the naturalness of human-machine collaboration and the safety of rehabilitation training. Attached Figure Description

[0062] Figure 1 This is a flowchart illustrating the control method for a lower limb rehabilitation exoskeleton robot in the first embodiment of the present invention.

[0063] Figure 2 This is a structural block diagram of the lower limb rehabilitation exoskeleton robot control system in the fourth embodiment of the present invention;

[0064] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0065] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0066] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0067] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0068] Example 1

[0069] Please see Figure 1 The figure shows a flowchart of the lower limb rehabilitation exoskeleton robot control method in the first embodiment of the present invention. As shown in the figure, the method includes the following steps:

[0070] In step S100, the patient's joint torque, lower limb joint angles, joint angular velocities, and plantar pressure data are collected through a joint encoder, a plantar pressure sensor, and an inertial measurement unit (IMU). Specifically, in this embodiment, after the patient wears the exoskeleton, the system collects the angles and angular velocities of the hip and knee joints in real time through the joint encoder, measures the plantar pressure distribution pressure, and acquires limb posture data through the IMU.

[0071] Step S200: Calculate rehabilitation status evaluation indicators based on joint torque, lower limb joint angle, and plantar pressure data. Specifically, the calculation expressions for the above rehabilitation status evaluation indicators are as follows:

[0072] ;

[0073] In the formula, V is the rehabilitation status evaluation index, and M is the joint torque. This is a torque reference value. R is the first weighting coefficient for joint torque, and R is the lower limb joint angle. For angle reference value, The second weighting coefficient for lower limb joint angles is denoted by S, where S represents plantar pressure data. This is a pressure reference value. This is the third weighting coefficient for plantar pressure data. By weighted fusion of joint torque, range of motion (lower limb joint angles), and support force (plantar pressure data), it simulates the therapist's comprehensive judgment of the patient's function. Using a baseline trajectory matching gait type as a template, the amplitude and offset are dynamically adjusted using the rehabilitation state V to adapt the trajectory to the patient's real-time ability.

[0074] Furthermore, the second and third weighting coefficients are dynamically adjusted based on gait phase variables, and the calculation expression for the adjusted weighting coefficients is as follows:

[0075] ;

[0076] ;

[0077] In the formula, This is the adjusted second weighting coefficient. This serves as the baseline value for the second weighting coefficient. This is the adjusted third weighting coefficient. This is the baseline value for the third weighting coefficient. For gait phase variables. To make it easy to understand, the gait phase interval, that is, the gait cycle, includes the support phase and the swing phase. When in the support phase, it is necessary to bear weight and maintain body stability, that is, to increase the weight of the third weight coefficient corresponding to the support force. When in the swing phase, it is necessary to swing the legs flexibly to complete the movement, that is, to increase the weight of the second weight coefficient corresponding to the lower limb joint angle.

[0078] Step S300: Collect the angle difference between the left and right hip joints, construct the gait phase variable representing the gait cycle, and input the time series composed of joint angular velocity and plantar pressure data into the pre-trained temporal classification model to identify the current gait type.

[0079] Specifically, the steps for constructing gait phase variables that characterize gait cycles include:

[0080] The angle difference between the left and right hip joints is low-pass filtered, and the filtered angle difference is normalized to an interval. Then, the phase variable is calculated by integration. The expression for calculating the gait phase variable is as follows:

[0081] ;

[0082] In the formula, For gait phase variables, For gait period, This is the angle value of the left hip joint. t represents the angle of the right hip joint;

[0083] The update expression for the gait period is:

[0084] ;

[0085] In the formula, For the updated gait cycle, Forgetting factor, For the previous state period, This represents the current gait cycle.

[0086] Transforming the integral of the angle difference between the left and right hip joints into a continuous phase variable essentially involves extracting gait cycle features through zero-crossing detection and normalization. Gait type recognition, i.e., the pre-trained temporal classification model, is the MiniRocket model. Its principle is to use random convolutional kernels to quickly extract local features of the time series and achieve high-precision pattern matching through a linear classifier.

[0087] Step S400: Match the desired trajectory under the corresponding gait type according to the gait type, and optimize the desired trajectory based on the rehabilitation status evaluation index to obtain the target optimized trajectory. Specifically, the system calls the corresponding baseline trajectory parameters from a preset database based on the identified gait type. The database is constructed based on the analysis of gait data from a large number of healthy people and includes key parameters under different movement modes. After obtaining the baseline trajectory, the rehabilitation status evaluation index is used as the core variable to perform real-time, dynamic, and personalized optimization of the trajectory. For example, for patients with weak muscle strength (low V value), the angular amplitude of joint movement can be reduced proportionally to reduce the difficulty of exercise and prevent muscle strain. As the rehabilitation progresses (V value increases), it is gradually increased to the normal range. Compared with traditional fixed trajectory training, dynamic adaptation can more effectively promote neural function remodeling, avoid overtraining or undertraining, and significantly improve rehabilitation efficiency.

[0088] Step S500: Establish a dynamic model of the exoskeleton coupled with the human body, and construct an acceleration constraint equation with the target optimized trajectory as the servo constraint target. Calculate the nominal torque based on the dynamic model parameters and the acceleration of the trajectory, and design a sliding surface by implementing monitoring of trajectory tracking error to generate a compensation torque, thereby calculating the control torque vector.

[0089] Preferably, in this embodiment, the calculation expression of the dynamic model is:

[0090] ;

[0091] In the formula, , , These represent the angle, angular velocity, and acceleration vectors of the joint, respectively. Represents the inertia matrix. Represents the matrix of Coriolis force and centripetal force. Represents the gravity term. This represents the control torque vector; specifically, the above... The inertial force term represents the torque necessary to accelerate the system. For example, to quickly lift a patient's lower leg, the motor must not only overcome the weight of the lower leg itself but also provide the force to accelerate it. (Inertia matrix) It includes information such as the mass and moment of inertia of each link; the above For the Coriolis force and centripetal force terms, when multiple joints move simultaneously, their velocities couple with each other, generating additional forces. This matrix describes the coupling effect of multi-joint motion. This is to balance the torque required to counteract the effects of gravity. For example, when a patient is standing, the hip and knee joints of the exoskeleton need to continuously output torque to resist gravity and prevent the patient from collapsing.

[0092] The expression for the constraint equation of acceleration is:

[0093] ;

[0094] In the formula, The constraint matrix represents the geometric relationship associated with the desired trajectory. This represents the constraint vector related to the velocity and acceleration of the trajectory.

[0095] Furthermore, the expression for calculating the control torque vector is:

[0096] ;

[0097] ;

[0098] ;

[0099] ;

[0100] ;

[0101] In the formula, This represents the nominal torque. Indicates expected acceleration, Indicates the desired speed. This indicates the compensation torque. Indicates sliding mode gain. Indicates the sliding surface. Represents the sliding mode control function. Indicates the reciprocal of the error. Indicates the parameters of the sliding surface. Indicates tracking error. Indicates the initial sliding mode gain. Indicates the learning rate. Let represent the sliding surface norm. Specifically, first, the deviation e between the actual trajectory and the desired trajectory is calculated. By designing the sliding surface s, the system state is attracted and maintained on the sliding surface s = 0, at which point the tracking error will asymptotically converge to zero. Based on the sliding surface, a compensation torque is designed, and an adaptive sliding control law is adopted. The sliding mode gain η can be adaptively adjusted according to the error norm. Compensation torque It is the feedback part of the controller. When it detects that the system deviates from the ideal trajectory, that is, s is not 0, it generates a torque to "pull" it back onto the right track.

[0102] Specifically, the desired trajectory is transformed into servo constraint equations, and the control torque is directly solved using generalized constraint forces, eliminating the need for Lagrange multipliers. Its robustness is achieved through sliding mode control, and the compensation torque suppresses model uncertainties through adaptive gain, ensuring that the tracking error is consistent and eventually bounded.

[0103] In some application scenarios of this embodiment, firstly, the rehabilitation goal (such as walking on flat ground) is transformed into a desired trajectory, which is expressed as the above-mentioned acceleration constraint equation. Then, using a known dynamic model, the control torque required to make the system both satisfy its own dynamic characteristics and accurately follow the constraint trajectory is analytically solved through the UK equation.

[0104] Preferably, in this embodiment, after obtaining the control torque vector, the method further includes:

[0105] The sensor data of joint angular velocity and angular acceleration are acquired in real time, the gait phase angle is calculated by the positive and negative tangent functions, and the results are defuzzified to the gait interval to match the gait cycle.

[0106] Based on the gait angle, the assist torque is calculated using a Gaussian kernel function. The assist amplitude coefficient is dynamically linked to rehabilitation status evaluation indicators. When the rehabilitation status evaluation indicators exceed preset values, the assist amplitude coefficient, which determines the maximum peak value of the assist torque, linearly decays to zero. Joint angular velocity and angular acceleration are acquired using an inertial measurement unit (IMU) or a joint encoder.

[0107] Furthermore, the calculation expression for the assist torque is as follows:

[0108] ;

[0109] ;

[0110] In the formula, To assist torque, The aforementioned assist amplitude coefficient determines the maximum peak value of the assist torque and is a core parameter for dynamic adjustment. For gait phase angle, The preset gait phase angle represents the gait phase point at which the assist torque should reach its peak (e.g., maximum leg lift assistance is required in the mid-swing phase). This value is calibrated based on a gait database of healthy individuals. To control the width parameter of the assist torque distribution around a preset gait phase angle, The smaller the value, the more concentrated the assistance. nearby; A larger value results in a smoother distribution, making it suitable for scenarios requiring a smooth transition. The joint angular velocity, The angular acceleration is used to simulate the peak distribution of joint moments in natural gait. The gait phase angle, by fusing angular velocity and angular acceleration, overcomes the limitations of a single parameter description, transforming interactive forces into trajectory fine-tuning quantities and achieving human-machine dynamic coupling.

[0111] To put it simply, the rehabilitation status evaluation index is a comprehensive quantitative indicator of a patient's ability, based on a weighted sum of joint torque, range of motion, and support force, where V ∈ [0, 1]. A higher V value indicates stronger voluntary movement ability. As the patient's ability improves (V increases), the assistance amplitude coefficient... The decay rate gradually decreases until it reaches zero when V=1. This achieves a smooth transition from "fully assisted" to "fully autonomous" care, preventing patients from developing device dependence and improving rehabilitation efficiency.

[0112] In summary, the lower limb rehabilitation exoskeleton robot control method in the above embodiments of the present invention, by synchronously collecting joint motion parameters, plantar pressure and inertial data, constructs a linkage relationship between rehabilitation status indicators and gait phase, accurately identifies gait type and generates personalized trajectory, and then decomposes nominal torque and robust compensation torque through UK (Udwadia-Kalaba) controller to achieve high-precision tracking while suppressing model uncertainty disturbances. Finally, it combines gait phase adaptive adjustment of the amplitude and timing of assist torque to form an "evaluation-planning-compensation" closed loop, effectively solving the problems of poor robustness of intent recognition, lack of flexibility in control rigidity, and disconnect between rehabilitation strategy and physiological timing in the prior art, significantly improving the naturalness of human-machine collaboration and the safety of rehabilitation training.

[0113] Example 2

[0114] A second embodiment of this application also provides a lower limb rehabilitation exoskeleton robot control system, which is used to implement the embodiments and preferred embodiments described herein, and will not be repeated hereafter. As used below, the terms "module," "unit," "subunit," etc., can refer to a combination of software and / or hardware that performs a predetermined function. Although the system described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0115] like Figure 2 As shown, the system includes: a data module 100, an evaluation module 200, a gait module 300, a trajectory module 400, and a control module 500.

[0116] Data module 100 is used to collect the patient's joint torque, lower limb joint angle, joint angular velocity and plantar pressure data through joint encoder, plantar pressure sensor and inertial measurement unit;

[0117] The evaluation module 200 is used to calculate rehabilitation status evaluation indicators based on joint torque, lower limb joint angle and plantar pressure data.

[0118] The gait module 300 is used to collect the angle difference between the left and right hip joints, construct gait phase variables that characterize the gait cycle, and input the time series of joint angular velocity and plantar pressure data into the pre-trained temporal classification model to identify the current gait type.

[0119] The trajectory module 400 is used to match the expected trajectory under the corresponding gait according to the gait type, and to optimize the expected trajectory based on the rehabilitation status evaluation index to obtain the target optimized trajectory.

[0120] The control module 500 is used to establish a dynamic model of the coupling between the exoskeleton and the human body, and to construct an acceleration constraint equation with the target optimized trajectory as the servo constraint target. Based on the dynamic model parameters and the acceleration of the trajectory, the nominal torque is calculated, and a sliding surface is designed by monitoring the trajectory tracking error to generate a compensation torque to calculate the control torque vector.

[0121] Preferably, in this embodiment, the gait module 300 is specifically used for:

[0122] The angle difference between the left and right hip joints is low-pass filtered, and the filtered angle difference is normalized to an interval. Then, the phase variable is calculated by integration. The expression for calculating the gait phase variable is as follows:

[0123] ;

[0124] In the formula, For gait phase variables, For gait period, This is the angle value of the left hip joint. t represents the angle of the right hip joint;

[0125] The update expression for the gait period is:

[0126] ;

[0127] In the formula, For the updated gait cycle, Forgetting factor, For the previous state period, This represents the current gait cycle.

[0128] Preferably, in this embodiment, the system further includes:

[0129] The adjustment module is used to acquire sensor data of joint angular velocity and angular acceleration in real time, calculate gait phase angle through tangent and inverse tangent functions, and deblur the results to the gait interval to match the gait cycle;

[0130] Based on the gait angle, the assist torque is calculated using a Gaussian kernel function. The assist amplitude coefficient is dynamically linked to the rehabilitation status evaluation index. When the rehabilitation status evaluation index is greater than the preset value, the assist amplitude coefficient that determines the maximum peak value of the assist torque is linearly reduced to zero.

[0131] It should be noted that the modules can be functional modules or program modules, and can be implemented in software or hardware. For modules implemented in hardware, the modules can reside in the same processor; or the modules can be located in different processors in any combination.

[0132] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A control system for a lower limb rehabilitation exoskeleton robot, characterized in that, include: The data module is used to collect the patient's joint torque, lower limb joint angle, joint angular velocity and plantar pressure data through joint encoders, plantar pressure sensors and inertial measurement units; The evaluation module is used to calculate rehabilitation status evaluation indicators based on joint torque, lower limb joint angle and plantar pressure data; The gait module is used to collect the angle difference between the left and right hip joints, construct gait phase variables that characterize the gait cycle, and input the time series of joint angular velocity and plantar pressure data into the pre-trained temporal classification model to identify the current gait type. The trajectory module is used to match the expected trajectory under the corresponding gait type according to the gait type, and to optimize the expected trajectory based on the rehabilitation status evaluation index to obtain the target optimized trajectory; The control module is used to establish a dynamic model of the coupling between the exoskeleton and the human body, and to construct an acceleration constraint equation with the target optimized trajectory as the servo constraint target. Based on the dynamic model parameters and the acceleration of the trajectory, the nominal torque is calculated, and a sliding surface is designed by monitoring the trajectory tracking error to generate a compensation torque in order to calculate the control torque vector. The calculation expression for the rehabilitation status evaluation index is as follows: ； In the formula, V is the rehabilitation status evaluation index, and M is the joint torque. This is a torque reference value. R is the first weighting coefficient for joint torque, and R is the lower limb joint angle. For angle reference value, The second weighting coefficient for lower limb joint angles is denoted by S, where S represents plantar pressure data. This is a pressure reference value. This is the third weighting coefficient for plantar pressure data; The gait module is specifically used for: The angle difference between the left and right hip joints is low-pass filtered, and the filtered angle difference is normalized to an interval. Then, the phase variable is calculated by integration. The expression for calculating the gait phase variable is as follows: ； In the formula, For gait phase variables, For gait period, This is the angle value of the left hip joint. t represents the angle of the right hip joint; The update expression for the gait period is: ； In the formula, For the updated gait cycle, Forgetting factor, For the previous state period, The current gait cycle; The calculation expression for the dynamic model is as follows: ； In the formula, , , These represent the angle, angular velocity, and acceleration vectors of the joint, respectively. Represents the inertia matrix. Represents the matrix of Coriolis force and centripetal force. Represents the gravity term. Represents the control torque vector; The expression for the constraint equation of acceleration is: ； In the formula, The constraint matrix represents the geometric relationship associated with the desired trajectory. Represents the constraint vector related to the velocity and acceleration of the trajectory; The expression for calculating the control torque vector is: ； ； ； ； ； In the formula, This represents the nominal torque. Indicates expected acceleration, Indicates the desired speed. This indicates the compensation torque. Indicates sliding mode gain. Indicates the sliding surface. This represents the sliding mode control function. Represents the reciprocal of the error. Indicates the parameters of the sliding surface. Indicates tracking error. Indicates the initial sliding mode gain. Indicates the learning rate. Denotes the sliding surface norm; The system also includes: The adjustment module is used to acquire sensor data of joint angular velocity and angular acceleration in real time, calculate gait phase angle through tangent and inverse tangent functions, and deblur the results to the gait interval to match the gait cycle; Based on the gait angle, the assist torque is calculated using a Gaussian kernel function. The assist amplitude coefficient is dynamically linked to the rehabilitation status evaluation index. When the rehabilitation status evaluation index is greater than the preset value, the assist amplitude coefficient that determines the maximum peak value of the assist torque is linearly reduced to zero.

2. The lower limb rehabilitation exoskeleton robot control system according to claim 1, characterized in that, The second and third weighting coefficients are dynamically adjusted based on gait phase variables, and the calculation expression for the adjusted weighting coefficients is as follows: ； ； In the formula, This is the adjusted second weighting coefficient. This serves as the baseline value for the second weighting coefficient. This is the adjusted third weighting coefficient. This is the baseline value for the third weighting coefficient. For gait phase variables.

3. The lower limb rehabilitation exoskeleton robot control system according to claim 1, characterized in that, The formula for calculating the assist torque is: ； ； In the formula, To assist torque, The assist amplitude coefficient, For gait phase angle, The preset gait phase angle, To control the width parameter of the assist torque distribution around a preset gait phase angle, The joint angular velocity, This is the joint angular acceleration.

Images