Intelligent assisting exoskeleton variable stiffness joint control method and device, equipment and medium

Abstract

This disclosure provides a method, device, equipment, and medium for controlling variable stiffness joints in an intelligent assisted exoskeleton. In some embodiments of this disclosure, the user's carrying weight and IMU data are acquired; the carrying weight is collected by a pressure sensor installed on the user's heel, and the IMU data is collected by an IMU installed on the user's lower leg; based on the carrying weight, a first stiffness value is quickly obtained by consulting a mapping table between the load range and the target stiffness; based on the IMU data, gait phase recognition is performed to obtain a second stiffness value; based on the first and second stiffness values, the motor rotation angle difference is calculated; and based on the motor rotation angle difference, the variable stiffness actuator is controlled. This disclosure improves the control accuracy and efficiency of the variable stiffness joint by using a pressure sensor and an IMU to acquire the first and second stiffness values, then calculating the motor rotation angle difference, and adjusting the output parameters of the variable stiffness actuator based on the motor rotation angle difference.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, and in particular to a method, device, equipment and medium for controlling the variable stiffness joints of an intelligent assisted exoskeleton. Background Technology

[0002] Mine emergency rescue operations are characterized by complex and highly dangerous environments. Rescuers often need to traverse narrow tunnels and rugged terrain for extended periods carrying heavy equipment such as respirators, life detectors, and demolition tools, with single loads reaching 20–40 kilograms. Under this high-intensity load, the waist and lower limb joints bear enormous stress, easily leading to muscle fatigue, strain, and even acute injury, severely restricting rescue efficiency and personnel safety. To improve human-machine collaboration and reduce physiological load, intelligent assistive exoskeletons for mining have become an important development direction for mine personal protective equipment in recent years. Among them, variable stiffness joint technology, due to its ability to dynamically balance support stiffness and movement flexibility, is considered a key path to improve the adaptability and energy efficiency of exoskeletons, especially suitable for the special working conditions of mines with variable loads and unstructured terrain.

[0003] The relevant technologies suffer from numerous problems, making it difficult to meet the actual needs of mining operations. Regarding equipment performance, rigid exoskeletons, due to their fixed high rigidity, not only result in inflexible movement but also cause severe human-machine interference when crossing obstacles, while consuming significant energy. Passive variable stiffness joints rely on mechanical pre-compression and cannot adapt to load changes in real time, limiting their application in complex working conditions. At the control level, electromyography (EMG) signals are susceptible to noise interference and have high processing delays, affecting the accuracy and timeliness of control.

[0004] Currently, variable stiffness joints have low control precision and low control efficiency. Summary of the Invention

[0005] This disclosure provides a method, device, equipment, and medium for controlling the variable stiffness joint of an intelligent assisted exoskeleton, which at least solves the problems of low control accuracy and low control efficiency of existing variable stiffness joints.

[0006] The technical solution disclosed herein is as follows: This disclosure provides a method for controlling the joints of an intelligent assisted exoskeleton with variable stiffness, including: The system acquires the human body's carrying weight and IMU data; wherein the carrying weight is acquired by a pressure sensor installed on the heel of the human's shoe, and the IMU data is acquired by an IMU installed on the lower leg of the human. Based on the load, the mapping table between the load range and the target stiffness is consulted to obtain the first stiffness value; Based on the IMU data, gait phase recognition is performed to obtain the second stiffness value; Calculate the motor rotation angle difference based on the first stiffness value and the second stiffness value; The variable stiffness actuator is controlled based on the motor rotation angle difference, wherein the variable stiffness actuator is installed at the ankle joint of the human body.

[0007] Optionally, the method further includes: If the current scene is detected to be an obstacle crossing scene, the joint stiffness of the variable stiffness actuator is adjusted to the target stiffness value.

[0008] Optionally, the method further includes: If the coronal plane tilt angle is detected to be greater than the set angle, the joint stiffness of the variable stiffness actuator is adjusted to the target stiffness value.

[0009] Optionally, the method further includes: The system acquires human motion-related parameters and system status-related parameters; wherein, the human motion-related parameters include: joint angles, angular velocities, and gait types; the system status-related parameters include: motor encoder angles, leaf spring force signals, and the exoskeleton's own status; Receive the current stiffness command and current torque command issued by the controller; The current stiffness command is mapped to the motor angle difference command, and combined with the current angle difference fed back by the encoder, the angle difference adjustment command is output through PID control and feedforward inertia compensation. By substituting the predicted human motion values ​​into the human-machine contact model, the total torque command is obtained.

[0010] Optionally, the method further includes: Human joint displacement signals are processed to calculate human motion compensation, friction compensation is calculated based on the Bowden rope friction model, and gravity compensation is superimposed. For fixed-step-size gait, PID control is used to adjust the residual error, and an RBF neural network is used to approximate the unmodeled error. For variable step length gait, the human-machine interaction force is converted into displacement compensation, which is then converted into compensation commands through a passive controller.

[0011] Optionally, the compensation instructions include: human motion compensation, friction compensation, gravity compensation, RBF neural network error compensation, and force compensation for admittance displacement conversion.

[0012] This disclosure also provides an intelligent assisted exoskeleton variable stiffness joint control device, comprising: The acquisition module is used to acquire the human body's carrying weight and IMU data; wherein, the carrying weight is acquired by a pressure sensor installed on the heel of the human's shoe, and the IMU data is acquired by an IMU installed on the lower leg of the human. The query module is used to query the mapping relationship table between the load range and the target stiffness based on the load weight, and obtain the first stiffness value; The recognition module is used to perform gait phase recognition based on the IMU data to obtain a second stiffness value; The calculation module is used to calculate the motor rotation angle difference based on the first stiffness value and the second stiffness value; The control module is used to control the variable stiffness actuator according to the motor rotation angle difference, wherein the variable stiffness actuator is installed at the ankle joint of the human body.

[0013] This disclosure also provides an electronic device, including: processor; Memory used to store processor-executable instructions; The processor is configured to execute instructions to implement the steps in the above method.

[0014] This disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0015] This disclosure also provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the method described above.

[0016] The technical solutions provided by the embodiments of this disclosure bring at least the following beneficial effects: In some embodiments of this disclosure, the carrying weight of the human body and IMU data are acquired. The carrying weight is collected by a pressure sensor installed at the heel of the human shoe, and the IMU data is collected by an IMU installed at the lower leg. Based on the carrying weight, a first stiffness value is quickly obtained by consulting a mapping table between the load range and the target stiffness. Based on the IMU data, gait phase recognition is performed to obtain a second stiffness value. Based on the first and second stiffness values, the motor angle difference is calculated. Based on the motor angle difference, a variable stiffness actuator is controlled, wherein the variable stiffness actuator is installed at the ankle joint. This disclosure improves the control accuracy and efficiency of the variable stiffness joint by using a pressure sensor and an IMU to acquire the first and second stiffness values, then calculating the motor angle difference, and adjusting the output parameters of the variable stiffness actuator based on the motor angle difference.

[0017] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.

[0019] Figure 1 A flowchart illustrating an intelligent assisted exoskeleton variable stiffness joint control method provided for an exemplary embodiment of this disclosure; Figure 2 This disclosure provides a schematic diagram of the structure of an intelligent assistive exoskeleton variable stiffness joint control device; Figure 3 A schematic diagram of the structure of an electronic device provided for an exemplary embodiment of this disclosure. Detailed Implementation

[0020] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0021] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure.

[0022] It should be noted that the user information involved in this disclosure includes, but is not limited to, user device information and user personal information; the collection, storage, use, processing, transmission, provision and disclosure of user information in this disclosure all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0023] To address the aforementioned technical problems, in some embodiments of this disclosure, the carrying weight and IMU data of the human body are acquired. The carrying weight is collected by a pressure sensor installed at the heel of the human shoe, and the IMU data is collected by an IMU installed at the lower leg. Based on the carrying weight, a first stiffness value is quickly obtained by consulting a mapping table between the load range and the target stiffness. Based on the IMU data, gait phase recognition is performed to obtain a second stiffness value. Based on the first and second stiffness values, the motor angle difference is calculated. Based on the motor angle difference, a variable stiffness actuator is controlled, wherein the variable stiffness actuator is installed at the ankle joint. This disclosure, by setting up pressure sensors and an IMU to acquire the first and second stiffness values, and then calculating the motor angle difference, adjusts the output parameters of the variable stiffness actuator based on the motor angle difference, thereby improving the control accuracy and efficiency of the variable stiffness joint.

[0024] The technical solutions provided by the embodiments of this disclosure are described in detail below with reference to the accompanying drawings.

[0025] Figure 1 This is a flowchart illustrating an exemplary embodiment of an intelligent assisted exoskeleton variable stiffness joint control method. Figure 1 As shown, the method includes: S101: Acquire the human body's carrying weight and IMU data; wherein, the carrying weight is acquired by a pressure sensor installed on the human's heel, and the IMU data is acquired by an IMU installed on the human's calf. S102: Based on the load weight, look up the mapping relationship table between the load range and the target stiffness to obtain the first stiffness value; S103: Based on IMU data, perform gait phase recognition to obtain the second stiffness value; S104: Calculate the motor rotation angle difference based on the first stiffness value and the second stiffness value; S105: The variable stiffness actuator is controlled according to the motor rotation angle difference. The variable stiffness actuator is installed at the ankle joint of the human body.

[0026] In this embodiment, the entity executing the above method can be a terminal device or a server.

[0027] The terminal device includes, but is not limited to, mobile stations (MS), mobile terminals, mobile phones, handsets, and portable equipment. This terminal device can communicate with one or more core networks via a radio access network (RAN). For example, the terminal device can be a mobile phone (or "cellular" phone), a computer with wireless communication capabilities, a computer with wireless transceiver capabilities, a virtual reality (VR) terminal device, an AR terminal device, a wireless terminal in industrial control, a wireless terminal in self-driving, a wireless terminal in remote medical care, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, a wireless terminal in a smart home, etc. The operating systems installed on the terminal device include, but are not limited to, iOS, Android, Windows, Linux, and Mac OS. In different networks, terminals may be called by different names, such as: user equipment, mobile station, user unit, station, cellular phone, personal digital assistant, wireless modem, wireless communication device, handheld device, laptop, cordless phone, wireless local loop station, television, etc. For ease of description, this embodiment will simply refer to it as terminal device.

[0028] In this embodiment, the implementation form of the server is not limited. For example, the server can be a conventional server, a cloud server, a cloud host, a virtual center, or other server devices. The server mainly consists of a processor, hard disk, memory, system bus, and other common computer architecture types.

[0029] In this embodiment, the human body's carrying weight and IMU data are acquired. The carrying weight is collected by a pressure sensor installed at the heel of the human's shoe, and the IMU data is collected by an IMU installed at the lower leg. Based on the carrying weight, a first stiffness value is quickly obtained by consulting a mapping table between the load range and the target stiffness. Based on the IMU data, gait phase recognition is performed to obtain a second stiffness value. Based on the first and second stiffness values, the motor angle difference is calculated. Based on the motor angle difference, a variable stiffness actuator is controlled, wherein the variable stiffness actuator is installed at the ankle joint. This disclosure improves the control accuracy and efficiency of the variable stiffness joint by setting up pressure sensors and IMUs to acquire the first and second stiffness values, and then calculating the motor angle difference. The output parameters of the variable stiffness actuator are adjusted based on the motor angle difference.

[0030] In some embodiments of this disclosure, the weight carried by the human body and IMU data are acquired. The weight carried is collected using a pressure sensor mounted on the heel of the foot, and the IMU data is collected using an IMU mounted on the lower leg. By placing sensors at corresponding locations on the lower limbs, embodiments of this disclosure can accurately collect pressure data and inertial measurement unit data.

[0031] In some embodiments of this disclosure, a first stiffness value is obtained by querying a mapping table between load range and target stiffness based on the load weight. The mapping table between load range and target stiffness is shown in Table 1 below. By setting corresponding target stiffness values ​​according to different load ranges, the exoskeleton can adaptively switch between high flexibility, balance mode, or high support mode.

[0032]

[0033] Table 1 In some embodiments of this disclosure, gait phase recognition is performed based on IMU data to obtain a second stiffness value; the motor rotation angle difference is calculated based on the first stiffness value and the second stiffness value. For example, firstly, acceleration and angular velocity data during the user's walking process are collected by the IMU, and a gait phase recognition algorithm (such as a threshold-based or machine learning method) is used to determine whether the user is currently in the support phase or the swing phase, thereby determining the corresponding second stiffness value (e.g., the support phase stiffness is 80 N·m / rad, and the swing phase stiffness is 30 N·m / rad); if the system's preset first stiffness value is 50 N·m / rad, then the difference between the two is calculated as ΔK = the second stiffness value. Based on the first stiffness value and the relationship between joint torque and stiffness τ = K·θ, the required adjustment angle difference Δθ = τ / ΔK of the motor can be calculated (assuming the torque τ is known).

[0034] In some embodiments of this disclosure, the variable stiffness actuator is controlled based on the motor rotation angle difference. The overall control flow is as follows: 1. Load-stiffness dynamic mapping mechanism.

[0035] By combining rescue action classifications (walking / climbing / obstacle crossing), multi-mode stiffness curves are preset to derive the relationship between joint angles and joint torques.

[0036] 2. Lightweight, high-strength variable stiffness joint actuator design.

[0037] Phase recognition algorithms can provide information on human motion in the current and future timeframes, while model predictive control algorithms can introduce future human motion information as a disturbance into the nonlinear physical model of a variable stiffness mechanism. In, (among them) The vector represents the system state (joint angle, angular velocity, etc.), and u represents the control input vector (motor output torque). This allows the physical model to make more accurate predictions about the future based on the current state, and the optimized calculation effectively compensates for the hysteresis and nonlinearity of the mechanical structure. The four-bar linkage and slider mechanism in this disclosure have nonlinear stiffness curves that are difficult to fit. To reduce the computational difficulty of model predictive control, the control of the preceding mechanism and the control of the subsequent elastic mechanism are separated, and will be described in turn below.

[0038] The front-end mechanism comprises two motor drives and a four-bar linkage. The control system is responsible for adjusting the contact force and contact position on the leaf spring to achieve stiffness adjustment. This disclosure primarily achieves the above objectives through a feedforward loop and PID control. Considering that the variable stiffness experiment is the core of the experimental requirements, stiffness control is the primary objective in the control objectives, while the total input torque of the leaf spring is a secondary control objective. The total torque Tr in the control design structure represents the desired torque input of the system, and its value is calculated by subsequent model predictive control methods. The stiffness curve value K is calculated from the load torque and stiffness command to obtain the motor rotation differential mode angle.

[0039] Without considering the force exerted by the four-bar linkage on the leaf spring, the nominal output torques of the two motors are respectively and Since the four-bar linkage is perfectly symmetrical, and the dynamic characteristics of the leaf spring and connecting rods are not considered, the total torque output by the leaf spring is... The distribution is evenly distributed to the transmission mechanisms on both sides. The control objective is that the force acting on the leaf spring is equal to the target total torque, i.e. Simultaneously, the differential angle also needs to be moved to a specified position. Ideally, in the output torque of the two motors, the common-mode component of the control torque is... This means that the sum of the output torques of the two motors is always equal to the total input torque command. Differential mode This is used to adjust the angle difference between the two motors, thereby adjusting the stiffness. For example... Increase at the same time Decrease At this time, it represents that motor 1 accelerates relative to motor 2, thereby increasing the angle difference. In actual control, the common-mode component... Much larger than those used to adjust joint angles Torque values. Furthermore, the inertia J1 and J2 of the two motor drives are not only different from the damping d1 and d2, but also very small compared to the common-mode torque. This results in the common-mode torque having a significant impact on the angle difference, severely interfering with the differential-mode torque's ability to control the angle difference; on the other hand, the control objective needs to satisfy the output... Inertia affects the actual torque value. Therefore, feedforward inertia compensation needs to be added at the output front end of each motor to improve the above-mentioned problem. Finally, considering that the current servo motor output torque characteristics are not perfectly linear, a gain circuit needs to be added at the final motor torque command input for manual adjustment during experimental testing. Here, we assume that the motor is an ideal linear motor and set the gain to 1.

[0040] In summary, the control flow is as follows: the stiffness command is first converted into the angle difference between the two motors. The angle difference command is compared with the difference in angle read by the encoders of the two motors. The difference result is adjusted by the PID controller, and torque compensation from the stiffness adjustment is added. Finally, the result is output to the motor for control.

[0041] Next, parameter configuration is performed. The main configurations in the control structure include establishing a mapping between torque and stiffness commands and the angle difference mode. Then, the motor drive feedforward compensation stage is configured, filter coefficients are set, and PID parameters are adjusted to control the angle difference. Finally, stiffness adjustment and torque compensation are configured. After configuration, simulation tests are conducted. For force servo applications, convergence is not possible without a load on the output. Therefore, the leaf spring is connected to the motor output, and the elastic mechanism output is fixed. The motor angle difference is set to a sinusoidal signal varying by 1Hz ± 0.2rad. A servo simulation experiment is performed on the torque, and the output torque value is determined by monitoring the force signal on the leaf spring.

[0042] In the experimental results, the tracked torque exhibited jitter at its maximum value. The main reason for this jitter was that, under large loads, the torque effect caused by changes in stiffness could not be completely offset by the feedforward method. However, it could still basically track the torque command signal, and the assist error was within the tolerance range of an assisted exoskeleton. The four-link angle tracking performance was good; apart from some jitter under heavy loads, it could still effectively converge and follow.

[0043] The next step is to verify the position control during freewheeling with the leaf springs removed. This is used to check the accuracy of the aforementioned part in model recognition. The four-bar linkage angle can track the sine command well, indicating that the model recognition is relatively accurate.

[0044] Although upper-level control can predict changes in joint angles and angular velocities, contact modeling of the human body is still necessary for the model to incorporate human motion into its predictions. Damping and springs are typically used to represent contact forces.

[0045] To incorporate human motion as a measurable disturbance into the control system, a general nonlinear model is used. Expand to Where u represents the control input, which is the total torque. d represents predictable external disturbances, specifically referring to the predicted human joint angles obtained in real time by the CPG algorithm in this system. and angular velocity prediction values Based on the spring-damped model, the human-machine contact torque The specific expression is:

[0046] in, and The angle of rotation between two links in the mechanism (unit: rad). The moment of inertia of the connecting rod (unit: kg·m²). These are the contact stiffness (unit: N·m / rad) and damping coefficient (unit: N·m·s / rad), respectively. The current stiffness is obtained by looking up a table. Assuming a certain period of time in the future... The target force control curve inside is The curve of additional force generated by human movement is To simplify matrix operations in subsequent model predictive control (MPC), the overall predictive output of the system is defined as:

[0047] It can also be expressed as a function of the system state x, denoted as ,in Let be the system's output function. To facilitate subsequent optimization calculations, a Taylor expansion is often performed at the current system time t, followed by forward Euler discretization and conversion to an incremental form. It is assumed that the joint angle and angular velocity information predicted by the upper-level control will be 0 after n time steps, and the control time domain is set to m, and the prediction time domain to p. From the discrete state-space equation above, the predicted output result for the next p steps can be obtained, as shown in the formula: ; The optimization objective is to minimize the torque tracking error over a future period of time; the second term represents the total torque. To minimize the incremental change, the optimization objective function is set as follows: ; in, , These represent the corresponding weight allocations. The input torque must be considered during the optimization process. The constraints are set as follows: ; Quadratic programming is a powerful tool for solving optimization problems. Rewriting the optimization objective yields the following QP configuration: ; The first solution obtained through QP optimization is used as the control input, and subsequent optimizations are performed repeatedly to achieve model predictive control. Control parameters need to be set for three aspects: control time, prediction time domain, control time domain, and measurable disturbance time domain. Since there is only a single objective, the weight allocation is 1. Based on model prediction experience, a step signal with a torque of 1 N·m is input to the model. The rise time is 6 ms, and the steady-state time is approximately 20 ms. Here, a control period of 2 ms is used, and the control time is 3 periods. The prediction time domain cannot exceed the prediction time domain of the entire model; here, 10 control times are temporarily used, and the entire system prediction also uses 10 prediction times.

[0048] The simulation section mainly includes two experiments. The first experiment tests the stiffness correction table, and the second experiment compares the impact of different disturbance signals on the prediction time-domain duration on the predictive control results. Both experiments are simulated using Matlab+Simulink. The output torque command signal is a 1Hz sine wave of 12 N·m, and the output stiffness command ranges from 60 to 150 N·m. Measurable human body disturbance and human joint angle signals are tested in two sets for comparison. The first set of stiffness values ​​is set to be only related to the common-mode difference of the motor angle. The second set uses a corrected stiffness table to compare the final torque output effect.

[0049] Without stiffness correction, the stiffness value of the elastic mechanism is too low under high load and too high under low torque load. This results in a low stiffness phenomenon near zero under low torque load, causing the motor output torque calculated by MPC to be much lower than the actual required torque, resulting in a brief pause in output torque at zero. In subsequent calculations, the actual high stiffness leads to a significant increase in output torque, causing the control system to jitter. After stiffness correction, the stiffness output torque pause near zero is effectively resolved, with jitter only occurring at the maximum torque. The reason for this is that while the stiffness value in the high stiffness region is also continuously increasing, it has little impact on the control results.

[0050] Furthermore, both experiments showed that the actual torque lagged behind the commanded torque. The reason for this is that the stiffness table ignores the hysteresis effect of mechanical components, making it difficult to compensate for the hysteresis in the actual situation within the control system. However, the main objective of this experiment remains to adjust the stiffness for related experiments.

[0051] The adaptive rhythm controller employs the following assistance strategies: First, the principle of the CPG phase recognition algorithm is introduced. Using perturbation theory, a first-order expansion of the CPG element is performed to derive the main formula determining phase convergence under the same frequency and waveform. This formula is then used to prove the convergence of the CPG element under complex limit cycles. By establishing a feedback structure to eliminate signal mean, the CPG phase recognition algorithm is constructed and verified using actual hip joint angle signals. Second, based on the idea of ​​common-mode torque compensation, the control block diagram of the pre-mechanism is constructed, and the model parameters of each inertial part are identified. Subsequently, a stiffness gradient table for variable stiffness torque compensation is constructed, and feedforward filtering and amplitude limiting output are set. The stiffness and torque accuracy are verified under Simulink-Adams co-simulation. Based on the model predictive control strategy, a simplified human-computer interaction model is used to establish the system state equation. Human motion is treated as a disturbance to establish the prediction matrix, and a quadratic programming optimization function is calculated. On this basis, a step signal is used to establish the system prediction and control time domain. Finally, the torque servo capability is verified through simulation.

[0052] 3. Active and passive impedance fusion control.

[0053] The active and passive impedance control strategy for exoskeletons first requires determining the impedance model of the system, simplifying the exoskeleton robot into a mass-damped-spring system, thereby establishing the relationship between the displacement of the exoskeleton and the input and output forces. Changing any parameter in the mass, damping, or spring system will cause a change in impedance, thereby controlling the change in the output human-machine interaction contact force.

[0054] set up This refers to error disturbances that still exist in the system after compensating for human movement and friction. To reduce these disturbances... To mitigate the impact on the control system, an RBF neural network is used for nonlinear compensation to improve the system's performance and accuracy. The dynamic model of the exoskeleton system after compensation is as follows:

[0055] In the formula: m is the equivalent mass of the motor. Human-computer interaction power Let k be the desired assist torque used in PID control, b be the equivalent joint stiffness, b be the equivalent joint damping, and f be the frictional force in the Bowden rope. This represents the linear displacement at the motor end. Let d represent the linear displacement of the knee joint, and d represent the unmodeled error of the system and external disturbances. and Let represent the second and first derivatives of the interaction force, respectively.

[0056] In some embodiments of this disclosure, when an obstacle-crossing scenario is detected, the joint stiffness of the variable stiffness actuator is adjusted to a target stiffness value. It should be noted that the embodiments of this disclosure do not limit the target stiffness value; the target stiffness value is the minimum stiffness value, for example, a target stiffness value of 28N. m / rad. If the current scene is detected as an obstacle crossing scenario, the joint stiffness of the variable stiffness actuator is adjusted to 28N. m / rad reduces the motion constraints of the mechanical structure, allowing the equipment to have more flexible posture adjustment capabilities when crossing obstacles, effectively reducing the obstruction of obstacles to the overall movement, and improving the smoothness and stability of passage.

[0057] In some embodiments of this disclosure, when a coronal plane tilt angle greater than a set angle is detected, the joint stiffness of the variable stiffness actuator is adjusted to a target stiffness value. It should be noted that the set angle is not limited in the embodiments of this disclosure and can be adjusted according to actual conditions. For example, the set angle is 30°. When a coronal plane tilt angle greater than 30° is detected, i.e., a fall is detected by the IMU, the joint stiffness of the variable stiffness actuator is adjusted to 28N. If the fall occurs in m / rad, immediately initiate emergency response procedures to minimize the damage caused by the fall.

[0058] It should be noted that the compensation instructions include: human motion compensation, friction compensation, gravity compensation, RBF neural network error compensation, and force compensation for admittance displacement conversion.

[0059] In some embodiments of this disclosure, human motion-related parameters and system state-related parameters are acquired. The human motion-related parameters include joint angles, angular velocities, and gait types. The system state-related parameters include motor encoder angles, leaf spring force signals, and the exoskeleton's own state. The system receives current stiffness and torque commands from the controller. The current stiffness command is mapped to a motor rotation angle difference command, and combined with the current rotation angle difference fed back by the encoder, a rotation angle difference adjustment command is output through PID control and feedforward inertia compensation. The predicted human motion values ​​are substituted into the human-machine contact model to obtain the total torque command. Human joint displacement signals are processed to calculate human motion compensation. Friction compensation is calculated based on the Bowden rope friction model and superimposed with gravity compensation. For fixed-step gait, PID control is used to adjust residual errors, and an RBF neural network is used to approximate unmodeled errors. For variable-step gait, human-machine interaction forces are converted into displacement compensation, which is then converted into compensation commands through a passive controller.

[0060] The overall control process of this disclosure is further explained below.

[0061] 1. Overall control chain relationship.

[0062] Upper-level decision-making (CPG phase recognition) → Lower-level control (including front-end mechanism control, MPC control, impedance / admittance control) → Actuator (variable stiffness mechanism + motor) → Human-machine interaction (output assistance).

[0063] This disclosure does not involve high-level decisions on "whether or not to provide assistance," but only addresses the implementation issue of "how to accurately and safely deliver assistance."

[0064] Core objective: To eliminate mechanical nonlinearity / hysteresis, compensate for interference, enable actuators to accurately respond to upper-level commands, and ensure safe human-machine interaction.

[0065] 2. The core input parameters that need to be obtained.

[0066] Human movement-related parameters: joint angles, angular velocity (provided by the CPG phase recognition algorithm, including current and future predictions), and gait type (fixed stride length / variable stride length, determined by the IMU sensor).

[0067] System status related factors: motor encoder angle (to calculate motor rotation angle difference), leaf spring load signal (to provide feedback on actual output torque), and exoskeleton's own status (gravity, Bowden rope friction, equivalent mass / damping / stiffness).

[0068] Command related: Stiffness command (target stiffness) and torque command (target assist magnitude) issued from the upper level.

[0069] 3. Core calculation process (divided into two main control modules).

[0070] (1) Variable stiffness actuator body control (pre-mechanism + MPC).

[0071] Stiffness adjustment calculation: The "target stiffness" is mapped to the motor angle difference command. Combined with the current angle difference fed back by the encoder, the motor control signal is output through PID adjustment and feedforward inertia compensation to realize the adjustment of leaf spring contact force / position (i.e. stiffness adjustment).

[0072] Precise torque control: The predicted human motion value (angle / angular velocity) is regarded as an interference and substituted into the human-machine contact model (spring-damping model). Through the quadratic programming optimization of MPC, the optimal total torque command is calculated to compensate for mechanical lag and nonlinearity and ensure torque tracking accuracy.

[0073] (2) Human-computer interaction optimization control (impedance / admittance compensation).

[0074] Impedance compensation calculation: Human joint displacement signals are processed by a high-order differential tracker to calculate the human motion compensation amount; friction compensation amount is calculated based on the Bowden rope friction model; gravity compensation is superimposed to reduce the interference of the exoskeleton's own impedance on the assist.

[0075] Fixed-step PID optimization: For fixed gait, the residual error is adjusted by PID, and then the unmodeled error (such as parameter deviation and external interference) is approximated by RBF neural network to further improve force control accuracy.

[0076] Variable step size admittance optimization: For variable gait, the human-machine interaction force is converted into displacement compensation, and then converted into force compensation command through a passive controller, reducing the ineffective interaction force caused by step size changes and ensuring safety.

[0077] 4. Final output (control commands).

[0078] Motor drive commands include torque commands (common mode, providing total assist) and angle difference adjustment commands (differential mode, adjusting stiffness).

[0079] Compensation commands: human motion compensation, friction compensation, gravity compensation, RBF neural network error compensation, and force compensation for admittance displacement conversion.

[0080] Figure 2 This disclosure provides a schematic diagram of the structure of an intelligent assisted exoskeleton variable stiffness joint control device 20, as shown in the exemplary embodiments. Figure 2 As shown, the intelligent assisted exoskeleton variable stiffness joint control device 20 includes: an acquisition module 21, a query module 22, an identification module 23, a calculation module 24, and a control module 25.

[0081] The acquisition module 21 is used to acquire the human body's carrying weight and IMU data; the carrying weight is acquired by a pressure sensor installed on the heel of the human's shoe, and the IMU data is acquired by an IMU installed on the lower leg of the human. The query module 22 is used to query the mapping relationship table between the load range and the target stiffness based on the load weight to obtain the first stiffness value; The recognition module 23 is used to perform gait phase recognition based on IMU data to obtain a second stiffness value; Calculation module 24 is used to calculate the motor rotation angle difference based on the first stiffness value and the second stiffness value; The control module 25 is used to control the variable stiffness actuator according to the motor rotation angle difference, wherein the variable stiffness actuator is installed at the ankle joint of the human body.

[0082] Optionally, the control module 25 can also be used for: If the current scene is detected as an obstacle crossing scene, the joint stiffness of the variable stiffness actuator will be adjusted to the target stiffness value.

[0083] Optionally, the control module 25 can also be used for: If the coronal plane tilt angle is detected to be greater than the set angle, the joint stiffness of the variable stiffness actuator will be adjusted to the target stiffness value.

[0084] Optionally, the control module 25 can also be used for: Acquire human motion-related parameters and system status-related parameters; among which, human motion-related parameters include: joint angles, angular velocities, and gait type; system status-related parameters include: motor encoder angles, leaf spring force signals, and exoskeleton status itself; Receive the current stiffness command and current torque command issued by the controller; The current stiffness command is mapped to the motor angle difference command, and combined with the current angle difference fed back by the encoder, the angle difference adjustment command is output through PID control and feedforward inertia compensation. By substituting the predicted human motion values ​​into the human-machine contact model, the total torque command is obtained.

[0085] Optionally, the control module 25 can also be used for: Human joint displacement signals are processed to calculate human motion compensation, friction compensation is calculated based on the Bowden rope friction model, and gravity compensation is superimposed. For fixed-step-size gait, PID control is used to adjust the residual error, and an RBF neural network is used to approximate the unmodeled error. For variable step length gait, the human-machine interaction force is converted into displacement compensation, which is then converted into compensation commands through a passive controller.

[0086] Optionally, the compensation instructions include: human motion compensation, friction compensation, gravity compensation, RBF neural network error compensation, and force compensation for admittance displacement conversion.

[0087] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0088] Figure 3 This is a schematic diagram of the structure of an electronic device provided as an exemplary embodiment of the present disclosure. For example... Figure 3 As shown, the electronic device includes a memory 31 and a processor 32. Additionally, the electronic device also includes a power supply component 33 and a communication component 34.

[0089] Memory 31 is used to store computer programs and can be configured to store various other data to support operation on the electronic device. Examples of this data include instructions for any application or method used to operate on the electronic device.

[0090] The memory 31 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0091] Communication component 34 is used for data transmission with other devices.

[0092] The processor 32 executes computer instructions stored in the memory 31 to: acquire the human body's carrying weight and IMU data; wherein the carrying weight is acquired by a pressure sensor installed at the heel of the human's shoe, and the IMU data is acquired by an IMU installed at the lower leg of the human; based on the carrying weight, quickly obtain a first stiffness value by querying a mapping table between the load range and the target stiffness; based on the IMU data, perform gait phase recognition to obtain a second stiffness value; calculate the motor angle difference based on the first stiffness value and the second stiffness value; control the variable stiffness actuator based on the motor angle difference, wherein the variable stiffness actuator is installed at the ankle joint of the human body; this disclosure improves the control accuracy and efficiency of the variable stiffness joint by setting pressure sensors and IMUs to acquire the first stiffness value and the second stiffness value, and then calculating the motor angle difference, and adjusting the output parameters of the variable stiffness actuator based on the motor angle difference.

[0093] Accordingly, embodiments of this disclosure also provide a computer-readable storage medium storing a computer program. When the computer-readable storage medium stores a computer program, and the computer program is executed by one or more processors, it causes one or more processors to perform... Figure 1 Each step in the method embodiment.

[0094] Accordingly, embodiments of this disclosure also provide a computer program product, which includes a computer program / instructions that are executed by a processor. Figure 1 Each step in the method embodiment.

[0095] The above Figure 3The communication component is configured to facilitate wired or wireless communication between the device containing the communication component and other devices. The device containing the communication component can access wireless networks based on communication standards, such as WiFi, 2G, 3G, 4G / LTE, 5G, or combinations thereof. In one exemplary embodiment, the communication component receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, the communication component also includes a Near Field Communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID), Infrared Data Association (IrDA) technology, Ultra-Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0096] The above Figure 3 The power supply component provides power to the various components of the device in which it resides. The power supply component may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device in which it resides.

[0097] The aforementioned electronic devices also include a display screen and audio components.

[0098] The display includes a screen, which may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors can sense not only the boundaries of touch or swipe actions, but also the duration and pressure associated with the touch or swipe operation.

[0099] An audio component may be configured to output and / or input audio signals. For example, the audio component includes a microphone (MIC) configured to receive external audio signals when the device containing the audio component is in an operating mode, such as call mode, recording mode, or voice recognition mode. The received audio signals may be further stored in memory or transmitted via a communication component. In some embodiments, the audio component also includes a speaker for outputting audio signals.

[0100] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0101] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0102] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0103] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0104] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0105] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0106] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0107] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0108] The above are merely specific embodiments of this disclosure, enabling those skilled in the art to understand or implement this disclosure. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to these embodiments, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for controlling the joint stiffness of an intelligent assisted exoskeleton, characterized in that, include: The system acquires the human body's carrying weight and IMU data; wherein the carrying weight is acquired by a pressure sensor installed on the heel of the human's shoe, and the IMU data is acquired by an IMU installed on the lower leg of the human. Based on the load, the mapping table between the load range and the target stiffness is consulted to obtain the first stiffness value; Based on the IMU data, gait phase recognition is performed to obtain the second stiffness value; Calculate the motor rotation angle difference based on the first stiffness value and the second stiffness value; The variable stiffness actuator is controlled based on the motor rotation angle difference, wherein the variable stiffness actuator is installed at the ankle joint of the human body.

2. The method according to claim 1, characterized in that, The method further includes: If the current scene is detected to be an obstacle crossing scene, the joint stiffness of the variable stiffness actuator is adjusted to the target stiffness value.

3. The method according to claim 1, characterized in that, The method further includes: If the coronal plane tilt angle is detected to be greater than the set angle, the joint stiffness of the variable stiffness actuator is adjusted to the target stiffness value.

4. The method according to claim 1, characterized in that, The method further includes: The system acquires human motion-related parameters and system status-related parameters; wherein, the human motion-related parameters include: joint angles, angular velocities, and gait types; the system status-related parameters include: motor encoder angles, leaf spring force signals, and the exoskeleton's own status; Receive the current stiffness command and current torque command issued by the controller; The current stiffness command is mapped to the motor angle difference command, and combined with the current angle difference fed back by the encoder, the angle difference adjustment command is output through PID control and feedforward inertia compensation. By substituting the predicted human motion values ​​into the human-machine contact model, the total torque command is obtained.

5. The method according to claim 1, characterized in that, The method further includes: Human joint displacement signals are processed to calculate human motion compensation, friction compensation is calculated based on the Bowden rope friction model, and gravity compensation is superimposed. For fixed-step-size gait, PID control is used to adjust the residual error, and an RBF neural network is used to approximate the unmodeled error. For variable step length gait, the human-machine interaction force is converted into displacement compensation, which is then converted into compensation commands through a passive controller.

6. The method according to claim 5, characterized in that, The compensation instructions include: human motion compensation, friction compensation, gravity compensation, RBF neural network error compensation, and force compensation for admittance displacement conversion.

7. A smart assisted exoskeleton variable stiffness joint control device, characterized in that, include: The acquisition module is used to acquire the human body's carrying weight and IMU data; wherein, the carrying weight is acquired by a pressure sensor installed on the heel of the human's shoe, and the IMU data is acquired by an IMU installed on the lower leg of the human. The query module is used to query the mapping relationship table between the load range and the target stiffness based on the load weight, and obtain the first stiffness value; The recognition module is used to perform gait phase recognition based on the IMU data to obtain a second stiffness value; The calculation module is used to calculate the motor rotation angle difference based on the first stiffness value and the second stiffness value; The control module is used to control the variable stiffness actuator according to the motor rotation angle difference, wherein the variable stiffness actuator is installed at the ankle joint of the human body.

8. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to execute instructions to implement the steps of the method as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-6.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1-6.

Images