Exoskeleton robot control method, device and equipment and storage medium

Abstract

The invention discloses a control method, device and equipment of an exoskeleton robot and a computer readable storage medium. The control method of the exoskeleton robot comprises the steps that kinematics parameters of interaction points of the robot and the environment are obtained; wherein the kinematics parameters comprise position parameters, speed parameters and acceleration parameters of interaction points of the robot and the environment; determining current environment rigidity, environment damping, robot rigidity and robot damping according to the kinematics parameters, predetermined environment rigidity, environment damping, robot rigidity and the incidence relation between the robot damping and the kinematics parameters; and performing motion control on the robot according to the environment rigidity, the environment damping, the robot rigidity and the robot damping in combination with a kinematics model met by the robot. According to the method, the impedance parameters can be dynamically and accurately adjusted in real time, so that the exoskeleton robot can improve the overall performance of the robot according to the environment change and the man-machine interaction state.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of robot control and human-computer interaction, and particularly relates to a control method, device and equipment of an exoskeleton robot and a computer readable storage medium. BACKGROUND

[0002] As a research hotspot in the field of rehabilitation engineering and wearable robots in recent years, the exoskeleton robot, especially the lower limb exoskeleton robot, is widely used in the rehabilitation training and walking assistance of patients with lower limb dysfunction. The exoskeleton robot needs to frequently and complexly interact with the human body and the external environment. In order to ensure the stability, safety and comfort of the system, the impedance control method is often used to make the robot have certain flexibility while ensuring the tracking performance, so as to realize natural human-computer interaction.

[0003] The traditional impedance control method usually sets fixed stiffness and damping parameters, but in actual application, the impedance characteristics of the environment have great uncertainty, such as the change of ground material, the difference of human behavior and other factors, which will affect the system performance. The fixed parameters cannot effectively adapt to the environmental disturbance, which may cause the problems of abnormal interaction force, decreased trajectory tracking accuracy, and even instability. SUMMARY

[0004] The purpose of the present application is to provide a control method, device and equipment of an exoskeleton robot and a computer readable storage medium, which can determine more accurate impedance parameters in real time, realize more accurate control of the exoskeleton robot, and improve the overall performance of the exoskeleton robot.

[0005] To solve the above technical problems, the present application provides a control method of an exoskeleton robot, comprising:

[0006] Obtaining kinematic parameters of an interaction point between the robot and the environment; wherein the kinematic parameters include position parameters, velocity parameters, acceleration parameters and interaction force measurement values of the interaction point between the robot and the environment;

[0007] According to the kinematic parameters, the pre-determined correlation between the environmental stiffness, the environmental damping, the robot stiffness, the robot damping and the kinematic parameters, the current environmental stiffness, the environmental damping, the robot stiffness and the robot damping are determined;

[0008] According to the environmental stiffness, the environmental damping, the robot stiffness and the robot damping, the kinematic model satisfied by the robot is combined to perform motion control on the robot.

[0009] In an optional embodiment of the present application, the process of determining the correlation between the environmental stiffness, the environmental damping and the kinematic parameters comprises:

[0010] According to the human-robot interaction force model satisfied by the robot , an interaction force conversion model is constructed and an interaction force estimation error formula is ; wherein, is an interaction force matrix, , is an environmental stiffness matrix and an environmental damping matrix respectively, is a position parameter matrix, is a velocity parameter matrix; is the measured value of the interaction force, is an interaction force estimation value, is an interaction force estimation error, is a regression vector, , is a position parameter, is a velocity parameter, is an environmental impedance parameter vector, , and is an environmental parameter estimation error, , is an environmental parameter estimation value; or 2, respectively, represent the parameters in two mutually perpendicular directions;

[0011] According to the interaction force estimation error formula , a prediction error function is constructed as wherein, , ; is an arbitrary constant, is an identity matrix;

[0012] According to the interaction force estimation error formula and the prediction error function, a composite learning law is constructed as to represent the relationship between the environmental stiffness and the environmental damping and the kinematic parameters; wherein, , are both set constants; is a projection function, ; , and , , =1 or 2.

[0013] In an optional embodiment of the present application, the process of determining the environmental stiffness and the environmental damping comprises:

[0014] obtaining the real value of the interaction force currently measured by the sensor;

[0015] According to the association between the human-robot interaction force real value, the interaction force estimation error and the prediction error function, the interaction force estimation value is optimized to determine the corresponding interaction force estimation value when the prediction error function converges;

[0016] According to the interaction force estimation value and the compound learning law, the environment stiffness and the environment damping corresponding to the prediction error function converging are determined.

[0017] In an optional embodiment of the present application, the process of determining the association between the robot stiffness and the robot damping and the kinematics parameters respectively comprises:

[0018] According to the human-robot interaction force model satisfied by the robot , a dynamics relationship is constructed as ; wherein, is a target position error, , is a position expectation value of the robot, , , ; is an interaction force expectation value;

[0019] According to the dynamics relationship, a cost function is constructed as ; wherein, are respectively set positive weights, ;

[0020] Based on the minimum value principle, the optimal solution corresponding to the minimum of the cost function is determined as ; wherein, ;

[0021] According to the optimal solution of the cost function and an expectation impedance model , a robot optimal stiffness is determined as , and the association between the robot stiffness and the kinematics parameters is represented by the robot optimal stiffness; wherein, , , respectively represent an expectation inertia matrix, an expectation stiffness matrix and an expectation damping matrix when the robot runs according to an expectation trajectory;

[0022] According to the human-robot interaction force model , the expectation impedance model is converted to obtain an expectation impedance conversion model as ;

[0023] According to the expectation impedance conversion model, the robot damping satisfies as the association between the robot damping and the kinematics parameters, wherein, to set the damping ratio.

[0024] In an alternative embodiment of the present application, the process of determining the robot stiffness and the robot damping comprises:

[0025] determining the robot stiffness according to the environmental stiffness and the environmental damping, in combination with the robot optimal stiffness

[0026] determining the robot damping according to the environmental stiffness, the environmental damping and the robot stiffness, in combination with the robot damping satisfying

[0027] A control device of an exoskeleton robot, comprising:

[0028] a data acquisition module for acquiring kinematics parameters of a robot and an environment interaction point; wherein the kinematics parameters include position parameters, velocity parameters and acceleration parameters of the robot and the environment interaction point;

[0029] a parameter operation module for determining current environmental stiffness, environmental damping, robot stiffness and robot damping according to the kinematics parameters, pre-determined association between the environmental stiffness, the environmental damping, the robot stiffness, the robot damping and the kinematics parameters;

[0030] a motion control module for performing motion control on the robot according to the environmental stiffness, the environmental damping, the robot stiffness and the robot damping, in combination with a kinematics model satisfied by the robot.

[0031] In an alternative embodiment of the present application, further comprising a first operation module for constructing an interaction force conversion model and an interaction force estimation error according to a human-robot interaction force model satisfied by the robot; is an interaction force matrix, , are an environmental stiffness matrix and an environmental damping matrix respectively, is a position parameter matrix, is a velocity parameter matrix; is a measured value of the interaction force, is an estimated value of the interaction force, is an interaction force estimation error, is a regression vector, , is a position parameter, is a velocity parameter,​​​ is an environmental impedance parameter vector, , and is an environmental parameter estimation error, , is an environmental parameter estimation value; or 2, respectively represent parameters in two mutually perpendicular directions; according to the interaction force estimation error , a prediction error function is constructed as , wherein, , ; is an arbitrary constant, is a unit matrix; according to the interaction force estimation error and the prediction error function, a composite learning law is constructed as , so that the composite learning rate represents the relationship between the environmental stiffness and the environmental damping and the kinematic parameters; wherein, , are both set constants; is a projection function, ; , and , , =1 or 2.

[0032] In an optional embodiment of the present application, a second operation module is further included, configured to construct a dynamic relationship as ; wherein, is a target position error, , is a position expectation value of the robot, , , , ; is an interaction force expectation value; according to the dynamic relationship, a cost function is constructed as ; wherein, are both set positive weights, ; based on the minimum value principle, the optimal solution corresponding to the minimum value of the cost function is determined as ; wherein, ; according to the optimal solution of the cost function and the expectation impedance model , the optimal stiffness of the robot is determined as , so that the optimal stiffness of the robot represents the relationship between the robot stiffness and the kinematic parameters; wherein, , , , respectively represent the expected inertia matrix, expected stiffness matrix and expected damping matrix when the robot runs according to the expected trajectory; according to the human-machine interaction force model The expected impedance model Perform the transformation to obtain the expected impedance transformation model: ; Determine the robot damping according to the desired impedance conversion model As the correlation relationship between the robot damping and the kinematic parameters, To set the damping ratio.

[0033] A control device for an exoskeleton robot, comprising:

[0034] memory for storing computer programs;

[0035] A processor is used to run the computer program to execute the steps of the control method of the exoskeleton robot as described in any one of the above items.

[0036] A computer-readable storage medium is used to store a computer program, wherein the computer program is executed to implement the steps of the control method of the exoskeleton robot as described in any one of the above items.

[0037] The present invention provides a control method, device, equipment and computer-readable storage medium for an exoskeleton robot. The control method for the exoskeleton robot includes obtaining kinematic parameters of the interaction point between the robot and the environment; wherein the kinematic parameters include position parameters, velocity parameters and acceleration parameters of the interaction point between the robot and the environment; determining the current environment stiffness, environment damping, robot stiffness and robot damping based on the kinematic parameters and the correlation between the predetermined environment stiffness, environment damping, robot stiffness and robot damping and the kinematic parameters; and performing motion control on the robot based on the environment stiffness, environment damping, robot stiffness and robot damping in combination with the kinematic model satisfied by the robot.

[0038] In this application, by predetermining the correlation between parameters such as environmental stiffness, environmental damping, robot stiffness and robot damping and the robot's kinematic parameters, the environmental stiffness, environmental damping, robot stiffness and robot damping that dynamically change with the robot's motion process can be determined in real time based on the robot's current kinematic parameters during the control of the exoskeleton robot's motion. That is, to a certain extent, the accuracy and reliability of the stiffness parameters and damping parameters in the robot's motion control process are improved, so that the exoskeleton robot can intelligently adjust its own control strategy according to environmental changes and human-computer interaction status, thereby improving the overall performance of the robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to make the technical scheme of the present application or prior art clearer, the accompanying drawings needed in the description of the embodiments or prior art will be briefly introduced. Obviously, the accompanying drawings described below only illustrate some of the embodiments of the present application, and all other embodiments obtained by a person of ordinary skill in the art without creative work on the basis of the accompanying drawings belong to the protection scope of the present application.

[0040] Figure 1 A flowchart of a control method of an exoskeleton robot provided by an embodiment of the present application.

[0041] Figure 2 A time-varying environmental stiffness curve determined by simulation of an embodiment in the present application.

[0042] Figure 3 A time-varying environmental damping curve determined by simulation of an embodiment in the present application.

[0043] Figure 4 A time-varying robot stiffness curve determined by simulation of an embodiment in the present application.

[0044] Figure 5 A time-varying robot damping curve determined by simulation of an embodiment in the present application.

[0045] Figure 6 A structural block diagram of a control device of an exoskeleton robot provided by an embodiment of the present application. DETAILED DESCRIPTION

[0046] The core of the present application is to provide a control method, device and equipment of an exoskeleton robot and a computer readable storage medium, which can more accurately dynamically adjust impedance parameters, thereby improving the control accuracy and overall performance of the robot.

[0047] In order to make the technical scheme of the present application or prior art clearer, the accompanying drawings needed in the description of the embodiments or prior art will be briefly introduced. Obviously, the accompanying drawings described below only illustrate some of the embodiments of the present application, and all other embodiments obtained by a person of ordinary skill in the art without creative work on the basis of the accompanying drawings belong to the protection scope of the present application.

[0048] As shown in Figure 1 , a flowchart of a control method of an exoskeleton robot provided by an embodiment of the present application. Figure 1

[0049] In a specific embodiment of the present application, the control method of the exoskeleton robot can include:

[0050] ​S11: Acquire kinematic parameters of the interaction point between the robot and the environment; wherein the kinematic parameters include position parameters, velocity parameters, acceleration parameters, and interaction force measurement values ​​of the interaction point between the robot and the environment;

[0051] S12: determining current environmental stiffness, environmental damping, robot stiffness, and robot damping according to the kinematic parameters, the predetermined correlation between the environmental stiffness, environmental damping, robot stiffness, and robot damping, and the kinematic parameters;

[0052] S13: Control the robot's motion according to the environment stiffness, environment damping, robot stiffness, and robot damping, combined with the kinematic model satisfied by the robot.

[0053] It can be understood that the position parameters, velocity parameters and acceleration parameters in the kinematic parameters of this embodiment can be determined based on the angle data measured by the encoder of the robot's own servo motor; and the interaction force measurement value is the force at the point of contact between the robot and the environment measured by the sensor. The environment referred to in this embodiment includes the human body and the ground that are in direct contact with the robot.

[0054] In this embodiment, by pre-establishing a determined correlation between environmental stiffness, environmental damping, robot stiffness, robot damping and kinematic parameters, and combining the currently acquired kinematic parameters, more accurate impedance parameters such as environmental stiffness, environmental damping, robot stiffness, robot damping at the current moment can be determined. Therefore, in the actual process of controlling the movement of the exoskeleton robot, more precise control of the movement of the exoskeleton robot can be achieved based on the current more accurate and reliable impedance parameters.

[0055] Based on the above discussion, the process of determining the relationship between environmental stiffness, environmental damping and kinematic parameters may include:

[0056] S21: Human-computer interaction force model based on robot satisfaction , building an interactive force conversion model and interaction force estimation error ;in, is the interaction force matrix, 、 are the environmental stiffness matrix and the environmental damping matrix respectively, is the location parameter matrix, is the velocity parameter matrix; is the interaction force measurement value, is the estimated value of the interaction force, is the interaction force estimation error, is the regression vector, , is a positional parameter, is a velocity parameter, is an environmental impedance parameter vector, , and is an environmental parameter estimation error, , is an environmental parameter estimation value; or 2, respectively represent parameters in two mutually perpendicular directions.

[0057] S22: According to the interaction force estimation error , a prediction error function is constructed as , wherein , ; is an arbitrary constant, is an identity matrix.

[0058] S23: According to the interaction force estimation error formula and the prediction error function, a composite learning law is constructed as to represent the relationship between the environmental stiffness and the environmental damping and the kinematic parameters with a composite learning rate; wherein , are both set constants; is a projection function, ; , and , , =1 or 2.

[0059] Based on the kinematic knowledge, the Cartesian space dynamics model of the robot is: ; wherein respectively represent the position, velocity and acceleration of the interaction point between the robot and the environment, that is, the above position parameters, velocity parameters and acceleration parameters; is the inertia matrix of the robot, and is a symmetric positive definite matrix, satisfying , are both constants greater than 0, is an arbitrary vector satisfying ; is the Coriolis force and centrifugal force matrix of the robot; is the gravity vector of the robot; is the friction force; is the control input; is the interaction force matrix, that is, representing the interaction force of the interaction point between the robot and the environment.

[0060] Further, the interaction force between the robot and the environment satisfies the human-robot interaction force model ; wherein , respectively, are environmental stiffness matrix and environmental damping matrix, and are symmetric matrices, that is , wherein, , and , are set upper and lower limit constants.

[0061] It should be noted that the in the embodiment is 1 or 2, which respectively represent parameters in two perpendicular directions, for example, when , , respectively represent components of environmental stiffness and environmental damping in the X-axis direction of the plane rectangular coordinate system; when , , respectively represent components of environmental stiffness and environmental damping in the Y-axis direction of the plane rectangular coordinate system. The concepts of the subsequent other parameters with as the subscript are similar, and will not be repeated hereinafter.

[0062] According to the above human-robot interaction force model, a human-robot interaction force model for a robot running according to an expected trajectory is further created as follows: ; wherein, is a position expected value of the robot; correspondingly, is a speed expected value of the robot, is an expected interaction force value; , are diagonal positive definite matrices, and are respectively expected stiffness matrix and expected damping matrix when the robot runs according to the expected trajectory.

[0063] In order to facilitate calculation, it is set that , is a regression vector, , is an environmental impedance parameter vector; it can be understood that each element in the environmental impedance parameter vector corresponds to the environmental stiffness and the environmental damping to be determined respectively; and should contain four elements, that is , respectively represent environmental stiffness in two mutually perpendicular directions, respectively represent environmental damping in two mutually perpendicular directions.

[0064] In addition, the environmental impedance parameter vector further satisfies:

[0065] ;

[0066] ;

[0067] .

[0068] It can be understood that, , are set boundary constants.

[0069] According to the setting of the above regression vector and the environmental impedance parameter vector, the human-machine interaction force model can be further converted into ; on this basis, further introduce the interaction force error formula , wherein, is the interaction force estimation error, is the interaction force measurement value, is the interaction force estimation value; combined with , the interaction force estimation error can be further determined; wherein, is the environmental parameter estimation error, is the environmental parameter estimation value, .

[0070] Further, according to the above human-machine interaction force estimation error, a prediction error function is constructed, , ; wherein T is a set time point, is any constant, is an identity matrix; obviously this prediction error function is a piecewise function.

[0071] Based on the above interaction force error and the above prediction error function, it can be obtained: and ; wherein, is the current time, is the starting time.

[0072] Therefore, a composite learning law can be further created ; wherein, , are set constants; is a projection function, , , and , , =1 or 2; it can be understood that the subscript here is used to distinguish whether it belongs to the stiffness parameter or the damping parameter.

[0073] As mentioned above, the prediction error function in the present application is a piecewise analysis function, and the reason why the above prediction error function is set as a piecewise function is to ensure in which is bounded in time and exponentially converges to 0 after time T.

[0074] Next, the piecewise prediction error function can guarantee In time T, the time range is bounded, and it is exponentially convergent to 0.

[0075] First, construct the Lyapunov function: .

[0076] The above composite learning law is brought into the time derivative of the Lyapunov function, and the following can be obtained:

[0077] ;

[0078] Under the premise of guaranteeing , if , .

[0079] Further, from , .

[0080] Therefore, combined with and , the ;

[0081] Therefore, when , at this time is bounded.

[0082] When , at this time exponentially converges to zero after time T.

[0083] Based on the above discussion, the elements of in the composite learning law determined in this embodiment are the environmental stiffness and the environmental damping, so the composite learning law can represent the relationship between the environmental stiffness and the environmental damping and the kinematic parameters.

[0084] Therefore, in the actual control process of the robot, the process of determining the environmental stiffness and the environmental damping of the robot at the current time can include:

[0085] S121: Obtain the real value of the interaction force measured by the sensor at the current time;

[0086] S122: According to the relationship between the real value of the human-robot interaction force, the interaction force estimation error, and the prediction error function, optimize the interaction force estimation value to determine the corresponding interaction force estimation value when the prediction error function converges.

[0087] S123: Determine the corresponding environment stiffness and environment damping when the prediction error function converges according to the interaction force estimation value and the composite learning law.

[0088] In this embodiment, by estimating the environmental impedance parameters (i.e. environmental stiffness and environmental damping) in real time, the robot can adaptively adjust its own impedance parameters in various complex interaction scenarios, enhancing the robustness and flexibility of the system. Moreover, the environmental impedance estimation uses a fusion error learning method, which improves the accuracy and real-time performance of modeling environmental disturbances compared to traditional single models.

[0089] Based on the above discussion, in another optional embodiment of the present application, the process of determining the robot stiffness and robot damping can include:

[0090] S31: According to the human-robot interaction force model satisfied by the robot , a dynamic relationship is constructed ; wherein, is the target position error, , is the position expectation value of the robot, , , ;

[0091] S32: According to the dynamic relationship, a cost function is constructed ; wherein, are the set positive weights, ;

[0092] Based on the minimum value principle, the optimal solution corresponding to the minimum value of the cost function is ; wherein, ;

[0093] S33: According to the optimal solution of the cost function and the expected impedance model , the optimal stiffness of the robot is determined as , which is the relationship between the robot stiffness and the kinematic parameters; wherein, , , , respectively represent the expected inertia matrix, the expected stiffness matrix and the expected damping matrix when the robot runs according to the expected trajectory;

[0094] S34: According to the human-robot interaction force model , the expected impedance model is converted to obtain the expected impedance conversion model as ;

[0095] S35: According to the impedance conversion model, the robot damping satisfies As the correlation between the robot damping and the kinematics parameters, wherein, to set the damping ratio.

[0096] Based on the above discussion, according to the human-robot interaction force model , and the human-robot interaction force model when the robot runs according to the expected trajectory , the dynamic relationship of the robot is determined as ; wherein, is the target position error, , is the position expectation value of the robot, , , .

[0097] According to the above dynamic relationship , in order to obtain the optimal stiffness of the robot in the interaction process, the cost function is constructed as ; wherein, are respectively set positive weights, . Obviously, under the condition of satisfying , the robot stiffness determined when the cost function is minimum is the optimal stiffness.

[0098] Therefore, based on the minimum value principle, the optimal solution of the cost function is solved, according to the Hamilton-Jacobi-Bellman (HJB) equation (the core tool for solving the infinite time optimal regulation problem in optimal control), the optimal control needs to satisfy:

[0099] ; wherein, is the value function and .

[0100] Taking the derivative of and setting it to 0, that is, , the optimal solution of the cost function is determined as .

[0101] In addition, the optimal solution obtained by solving the cost function is substituted into the Hamilton-Jacobi-Bellman equation, that is:

[0102] ;

[0103] Based on the above formula, it can be deduced that ; in the case that is not 0, that is, further is determined.

[0104] Convert to , and solve the quadratic equation with respect to , so as to determine .

[0105] According to the human-robot interaction force model running according to the expected trajectory, the expected impedance model of the robot is further constructed ; wherein are all diagonal positive definite matrices, that is , , , respectively representing the expected inertia matrix, the expected stiffness matrix and the expected damping matrix of the robot running according to the expected trajectory; and and each element in the robot stiffness and robot damping required to be determined by the present application.

[0106] In the case of steady-state operation of the robot, , , the , is substituted into the expected impedance model , so as to obtain: .

[0107] And in combination with , the .

[0108] In addition, further according to , , , ; thus, can be further converted to ;

[0109] According to the above , it can be determined that:

[0110] ;

[0111] Thus, the optimal stiffness of the robot is determined, that is, the relationship between the robot stiffness and the kinematic parameters is .

[0112] On this basis, further according to the human-robot interaction force model , the expected impedance model is converted to obtain the expected impedance conversion model ; thus the scalar form of the expected impedance conversion model is .

[0113] Further, convert to the form of a second-order system: ; wherein , , is set to the damping ratio.

[0114] According to the above, the robot stiffness and the robot damping determined in the embodiment are respectively determined according to the kinematic parameters. .

[0115] As described above, the correlations between the robot stiffness and the robot damping determined in the embodiment and the kinematic parameters are essentially the correlations between the robot stiffness and the robot damping and the environmental stiffness and the environmental damping; thus, in actual application, as long as the environmental stiffness and the environmental damping are determined according to the kinematic parameters, the robot stiffness at the current moment can be further determined according to the environmental stiffness, the environmental damping and the robot stiffness, and the robot damping is determined according to the robot stiffness and the robot damping.

[0116] In the embodiment, the compound error and the performance function are jointly optimized, the interactive force is suppressed and the trajectory tracking accuracy is considered, and the naturalness and comfort of human-robot interaction are improved.

[0117] In summary, in the present application, the correlations between the parameters such as the environmental stiffness, the environmental damping, the robot stiffness and the robot damping and the kinematic parameters of the robot are determined in advance, so that the environmental stiffness, the environmental damping, the robot stiffness and the robot damping dynamically changing with the movement of the robot can be determined in real time based on the kinematic parameters of the robot during the movement control of the exoskeleton robot, that is, the accuracy and reliability of the stiffness parameters and the damping parameters in the movement control process of the robot are improved to some extent, so that the exoskeleton robot can intelligently adjust its control strategy according to the environmental changes and the human-robot interaction state, thereby improving the overall performance of the robot.

[0118] Based on the above embodiment, referring to Figures 2 to 5 , Figure 2 is the environmental stiffness curve varying with time determined in the simulation of the embodiment in the present application; Figure 3 is the environmental damping curve varying with time determined in the simulation of the embodiment in the present application; Figure 4 is the robot stiffness curve varying with time determined in the simulation of the embodiment in the present application; Figure 5 is the robot damping curve varying with time determined in the simulation of the embodiment in the present application.

[0119] Take , , , , , the adaptive simulation of the environmental stiffness and the environmental damping is performed, and the simulation results are as follows​Figure 2 and Figure 3 shown; based on Figure 2 and Figure 3 It can be seen that before 5 seconds, the values ​​of environmental stiffness and environmental damping vary greatly. After 5 seconds, the estimated environmental stiffness and environmental damping converge and are very close to their actual values, that is, , .

[0120] Furthermore, based on the environmental stiffness and environmental damping, , , ;like Figure 4 and Figure 5 As shown in FIG, after 5 seconds, both the robot stiffness and the robot damping converge, and the results converge to the optimal stiffness and damping obtained by minimizing the cost function of the robot motion and interaction force.

[0121] The following is an introduction to the control device of the exoskeleton robot provided by an embodiment of the present invention. The control device of the exoskeleton robot described below and the control method of the exoskeleton robot described above can be referenced to each other.

[0122] Figure 6 The structural block diagram of the control device of the exoskeleton robot provided by the embodiment of the present invention is shown in FIG. Figure 6 The control device of the exoskeleton robot may include:

[0123] The data acquisition module 100 is used to obtain kinematic parameters of the interaction point between the robot and the environment; wherein the kinematic parameters include position parameters, velocity parameters and acceleration parameters of the interaction point between the robot and the environment;

[0124] a parameter calculation module 200 for determining current environmental stiffness, environmental damping, robot stiffness, and robot damping according to the kinematic parameters, the predetermined correlation between environmental stiffness, environmental damping, robot stiffness, and robot damping, and the kinematic parameters;

[0125] The motion control module 300 is configured to perform motion control on the robot according to the environment stiffness, the environment damping, the robot stiffness and the robot damping, in combination with a kinematic model satisfied by the robot.

[0126] In an optional embodiment of the present application, a first operation module is further included for calculating the human-machine interaction force model satisfied by the robot. , building an interactive force conversion model The error formula for estimating the interaction force is: ;in, is the interaction force matrix, 、 are the environmental stiffness matrix and the environmental damping matrix respectively, is the location parameter matrix, is the velocity parameter matrix; is the interaction force measurement value, is the estimated value of the interaction force, is the interaction force estimation error, is the regression vector, , is a positional parameter, is the speed parameter, is the environmental impedance parameter vector, ,and is the environmental parameter estimation error, , is the estimated value of the environmental parameter; or 2, respectively representing the parameters in two mutually perpendicular directions; according to the interaction force estimation error formula: , construct the prediction error function as ,in, , ; is any constant, is the unit matrix; according to the interaction force estimation error formula and prediction error function, the composite learning law is constructed as , the relationship between environmental stiffness and environmental damping and kinematic parameters is characterized by compound learning rate; among them, 、 All are set constants; is the projection function, ; ,and , , =1 or 2.

[0127] In an optional embodiment of the present application, the parameter calculation module 200 is specifically used to obtain the true value of the interaction force currently measured by the sensor; based on the true value of the human-computer interaction force, the correlation relationship between the interaction force estimation error and the prediction error function, the interaction force estimation value is optimized to determine the interaction force estimation value corresponding to when the prediction error function converges; based on the interaction force estimation value and the composite learning law, the environmental stiffness and the environmental damping corresponding to when the prediction error function converges are determined.

[0128] In an optional embodiment of the present application, a second operation module is further included for calculating the human-machine interaction force model satisfied by the robot. , the dynamic relationship is constructed as ;in, is the target position error, , a position expectation value of the robot, , , ; constructing a cost function according to the dynamic relationship ; wherein, respectively set positive weights, ; determining, based on a minimum principle, that an optimal solution corresponding to a minimum of the cost function is ; wherein, ; determining, according to the optimal solution of the cost function and an expected impedance model , an optimal stiffness of the robot as characterizing a relationship between the robot stiffness and the kinematic parameters with the optimal stiffness of the robot; wherein, , , respectively represent an expected inertia matrix, an expected stiffness matrix and an expected damping matrix when the robot runs according to an expected trajectory; converting the expected impedance model according to the human-robot interaction force model to obtain an expected impedance conversion model as ; determining, according to the expected impedance conversion model, that the robot damping satisfies as a relationship between the robot damping and the kinematic parameters, wherein, is a set damping ratio.

[0129] In an optional embodiment of the present application, the parameter operation module 200 is specifically configured to determine the robot stiffness according to the environment stiffness and the environment damping, in combination with the optimal stiffness of the robot ; determine the robot damping according to the environment stiffness, the environment damping and the robot stiffness, in combination with the robot damping satisfying .

[0130] The control device of the exoskeleton robot of the present embodiment is used to implement the control method of the exoskeleton robot as described above, and therefore the specific embodiments of the control device of the exoskeleton robot can be seen in the embodiment part of the control method of the exoskeleton robot in the foregoing, and the specific embodiments can be referred to the description of the corresponding embodiment part, which will not be repeated here.

[0131] The present application also provides an embodiment of a control device of an exoskeleton robot, which comprises:

[0132] a memory for storing a computer program;

[0133] a processor for running the computer program to perform the steps of implementing the control method of the exoskeleton robot as described in any of the above.

[0134] The processor executes steps of a control method of the exoskeleton robot implemented by a computer program, comprising:

[0135] Obtaining kinematic parameters of the robot and the environment interaction point; wherein the kinematic parameters comprise position parameters, velocity parameters, acceleration parameters and interaction force measurement values of the robot and the environment interaction point; determining current environment stiffness, environment damping, robot stiffness and robot damping according to the kinematic parameters, a pre-determined correlation between the kinematic parameters, the environment stiffness, the environment damping, the robot stiffness and the robot damping; and performing motion control on the robot according to the environment stiffness, the environment damping, the robot stiffness and the robot damping in combination with a kinematic model satisfied by the robot.

[0136] The present application also provides a computer readable storage medium for storing a computer program, wherein the computer program is executed to implement steps of the control method of the exoskeleton robot according to any one of the above.

[0137] The computer readable storage medium comprises random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, register, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0138] It should be noted that, in this document, the relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or sequence between the entities or operations. Moreover, the terms “include”, “contain” or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device inherently includes a series of elements. Without more limitations, the element defined by the statement “includes a” does not exclude the presence of another identical element in the process, method, article or device including the element. In addition, the above technical solutions provided by the embodiments of the present application have not been described in detail, so as not to be too verbose, which are consistent with the implementation principles of the corresponding technical solutions in the prior art.

[0139] The principles and implementation modes of the present application are described by using specific examples in this document, and the above description of the embodiments is only used to help understand the method and its core idea of the present application. It should be noted that, for those skilled in the art, without departing from the principles of the present application, some improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the present application.

Claims

1. A control method for an exoskeleton robot, characterized in that: include: Acquiring kinematic parameters of the interaction point between the robot and the environment; wherein the kinematic parameters include position parameters, velocity parameters, acceleration parameters and interaction force measurement values ​​of the interaction point between the robot and the environment; Determining current environmental stiffness, environmental damping, robot stiffness, and robot damping according to the kinematic parameters, a correlation relationship between the predetermined environmental stiffness, environmental damping, robot stiffness, and robot damping and the kinematic parameters; The robot is motion-controlled according to the environment stiffness, the environment damping, the robot stiffness, and the robot damping in combination with a kinematic model satisfied by the robot.

2. The control method of the exoskeleton robot according to claim 1, wherein: The process of determining the correlation between the environmental stiffness, the environmental damping and the kinematic parameters includes: According to the human-computer interaction force model satisfied by the robot , building an interactive force conversion model The error formula for estimating the interaction force is: ;in, is the interaction force matrix, 、 are the environmental stiffness matrix and the environmental damping matrix respectively, is the location parameter matrix, is the velocity parameter matrix; is the interaction force measurement value, is the estimated value of the interaction force, is the interaction force estimation error, is the regression vector, , is a positional parameter, is the speed parameter, is the environmental impedance parameter vector, ,and is the environmental parameter estimation error, , is the estimated value of the environmental parameter; or 2, respectively representing parameters in two mutually perpendicular directions; According to the interaction force estimation error formula: , construct the prediction error function as ,in, , ; is any constant, is the identity matrix; According to the interaction force estimation error formula and the prediction error function, a composite learning law is constructed as follows: , the relationship between the environmental stiffness and the environmental damping and the kinematic parameters is characterized by the compound learning rate; wherein, 、 All are set constants; is the projection function, ; ,and , , 1 or 2.

3. The control method of the exoskeleton robot according to claim 2, characterized in that: The process of determining the environmental stiffness and the environmental damping includes: Get the actual value of the interaction force currently measured by the sensor; Optimizing the estimated interaction force value according to the correlation between the true value of the human-computer interaction force, the estimated interaction force error, and the prediction error function to determine the estimated interaction force value corresponding to when the prediction error function converges; The environmental stiffness and the environmental damping corresponding to when the prediction error function converges are determined according to the interaction force estimation value and the composite learning law.

4. The control method of the exoskeleton robot according to claim 2 or 3, characterized in that: The process of determining the correlation between the robot stiffness and the robot damping and the kinematic parameters respectively includes: According to the human-computer interaction force model satisfied by the robot , the dynamic relationship is constructed as ;in, is the target position error, , is the expected position of the robot, , , ; is the expected value of interaction force; According to the dynamic relationship, the cost function is constructed ;in, Set positive weights for ; Based on the minimum principle, the optimal solution corresponding to the minimum cost function is determined to be ;in, ; According to the optimal solution of the cost function and the expected impedance model , determine the optimal stiffness of the robot as , the correlation between the robot stiffness and the kinematic parameters is characterized by the optimal stiffness of the robot; wherein, 、 、 , respectively represent the expected inertia matrix, expected stiffness matrix and expected damping matrix when the robot runs along the expected trajectory; According to the human-computer interaction model The expected impedance model Perform the transformation to obtain the expected impedance transformation model: ; Determine the robot damping according to the expected impedance conversion model As the correlation relationship between the robot damping and the kinematic parameters, To set the damping ratio.

5. The control method of the exoskeleton robot according to claim 4, characterized in that: The process of determining the robot stiffness and the robot damping includes: According to the environmental stiffness and the environmental damping, the robot's optimal stiffness is combined , determining the stiffness of the robot; According to the environment stiffness, the environment damping and the robot stiffness, combined with the robot damping to meet , determine the robot damping.

6. A control device for an exoskeleton robot, characterized in that: include: A data acquisition module, configured to acquire kinematic parameters of the interaction point between the robot and the environment; wherein the kinematic parameters include position parameters, velocity parameters, and acceleration parameters of the interaction point between the robot and the environment; a parameter calculation module, configured to determine current environmental stiffness, environmental damping, robot stiffness, and robot damping according to the kinematic parameters, a correlation relationship between the predetermined environmental stiffness, environmental damping, robot stiffness, and robot damping and the kinematic parameters; A motion control module is used to control the motion of the robot according to the environmental stiffness, the environmental damping, the robot stiffness and the robot damping, in combination with a kinematic model satisfied by the robot.

7. The control device for the exoskeleton robot according to claim 6, wherein: It also includes a first computing module for calculating the human-machine interaction force model satisfied by the robot. , building an interactive force conversion model and interaction force estimation error ;in, is the interaction force matrix, 、 are the environmental stiffness matrix and the environmental damping matrix respectively, is the location parameter matrix, is the velocity parameter matrix; is the interaction force measurement value, is the estimated value of the interaction force, is the interaction force estimation error, is the regression vector, , is a positional parameter, is the speed parameter, is the environmental impedance parameter vector, ,and is the environmental parameter estimation error, , is the estimated value of the environmental parameter; or 2, respectively representing the parameters in two mutually perpendicular directions; according to the interaction force estimation error , construct the prediction error function as ,in, , ; is any constant, is the unit matrix; the error is estimated according to the interaction force And the prediction error function, construct the composite learning law as , the relationship between the environmental stiffness, the environmental damping and the kinematic parameters is characterized by the composite learning law; wherein, 、 All are set constants; is the projection function, ; ,and , , =1 or 2.

8. The control device for the exoskeleton robot according to claim 7, wherein: It also includes a second computing module for calculating the human-machine interaction force model satisfied by the robot. , the dynamic relationship is constructed as ;in, is the target position error, , is the expected position of the robot, , , ; is the expected value of the interaction force; according to the dynamic relationship, the cost function is constructed ;in, Set positive weights for Based on the minimum principle, the optimal solution corresponding to the minimum cost function is determined to be ;in, ; According to the optimal solution of the cost function and the expected impedance model , determine the optimal stiffness of the robot as , the correlation between the robot stiffness and the kinematic parameters is characterized by the optimal stiffness of the robot; wherein, 、 、 , respectively represent the expected inertia matrix, expected stiffness matrix and expected damping matrix when the robot runs according to the expected trajectory; according to the human-machine interaction force model The expected impedance model Perform the transformation to obtain the expected impedance transformation model: ; Determine the robot damping according to the desired impedance conversion model As the correlation relationship between the robot damping and the kinematic parameters, To set the damping ratio.

9. A control device for an exoskeleton robot, characterized in that: include: memory for storing computer programs; A processor is configured to run the computer program to execute the steps of the exoskeleton robot control method according to any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium is used to store a computer program, and the computer program is executed to implement the steps of the control method of the exoskeleton robot according to any one of claims 1 to 5.

Images