Output feedback robust control method, system and equipment for humanoid robot

Abstract

The invention provides an output feedback robust control method, system and device for a humanoid robot, and belongs to the technical field of humanoid robot control. The method comprises the following steps: establishing a kinetic model of a robot limb end system, and defining a state variable, a control variable and an output variable; determining a first robust control equation based on a state feedback optimal control principle, the kinetic model and an output feedback controller constructed by the output variables; converting the first robust control equation into a second robust control equation based on a mapping relation between a state variable and an output variable in the kinetic model, determining a to-be-estimated parameter vector according to the second robust control equation, and performing online estimation based on a preset adaptive law to obtain a control gain matrix; a second robust control equation is constructed based on the output variables; and based on the control gain matrix and the output variable obtained in real time, generating a corresponding control variable to be applied to the robot acra joint so as to carry out robust control on the robot acra system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of humanoid robot control, and in particular to an output feedback robust control method, system and device for a humanoid robot. BACKGROUND

[0002] With the development of the times and the progress of science and technology, humanoid robots have entered the prototype testing and trial application stage. With the continuous increase of research and development investment in this field, the intelligence level of humanoid robots is rising. At the same time, the core of robot limb control also needs more advanced algorithms.

[0003] The existing humanoid robot system is a complex system with multiple joints, strong coupling and high nonlinearity. Using a model-dependent control method often requires a large number of system parameters, and using a model-free control method with poor interpretability has the problem of a long training process. Common control algorithms, such as PID control, adaptive control, model predictive control, etc., the PID control algorithm has low modeling requirements for the system and is widely used in industrial robot systems. However, for humanoid robots, the effectiveness of traditional control strategies represented by PID is limited, and in the face of sudden external force collision, road disturbance and other scenes, the anti-interference ability is weak, and the system state recovers slowly after being disturbed.

[0004] Therefore, in view of the complex dynamic characteristics of humanoid robots, breaking through the limitations of existing methods in model dependence, anti-interference ability, control efficiency, etc. has become the key to promoting the progress of precise robust control technology. SUMMARY

[0005] The embodiments of the present application provide an output feedback robust control method, system and device for a humanoid robot, which is used to solve the technical problem of how to overcome the limitations of model dependence, anti-interference ability, control efficiency, etc. for a system with complex dynamic characteristics of humanoid robots, and to perform precise output feedback robust control.

[0006] In one aspect, the embodiments of the present application provide an output feedback robust control method for a humanoid robot, which comprises: establishing a dynamics model of a robot limb end system, and defining state variables, control variables and output variables; wherein the dynamics model is constructed based on Lagrange equation, and at least includes an inertia matrix, a Coriolis force vector, a gravity vector and a disturbance torque; the state variable is a motion state parameter of a limb end joint, the control variable is a control input parameter acting on the limb end joint, and the output variable is a measurable limb end joint state parameter; determine a first robust control equation based on the state feedback optimal control principle, the dynamic model, and an output feedback controller constructed based on the output variable; wherein the first robust control equation is added with a preset state weight matrix and a control strategy weight matrix to adjust the control performance; convert the first robust control equation into a second robust control equation based on a mapping relationship between the state variable and the output variable in the dynamic model, to determine a to-be-estimated parameter vector according to the second robust control equation, and to obtain a control gain matrix in the output feedback controller based on a preset adaptive law and online estimation; generate a corresponding control variable based on the control gain matrix and the output variable acquired in real time, and apply the control variable to a robot limb end joint, to perform robust control on the robot limb end system.

[0007] In an implementation manner of the present application, the method is applied to output feedback robust control of a lower limb of a humanoid robot; the robot limb end system comprises a plurality of lower limb joints with one or more degrees of freedom; the lower limb joints at least comprise a crotch joint, a hip joint, a leg joint, a knee joint, an ankle joint, and a waist joint; establish a dynamic model of the robot limb end system, specifically comprising: model a kinematics equation of the lower limb of the humanoid robot based on a pose vector corresponding to each of the lower limb joints, to obtain a motion equation; establish the dynamic model based on the motion equation and the Lagrange equation.

[0008] In an implementation manner of the present application, the state variable is defined as: ; wherein, , represents a position coordinate of a lower limb joint, represents a pose of the lower limb of the humanoid robot in a world coordinate system, represents a pose vector of a limb end joint, is a joint velocity vector, is a first variable in the state variable, is a second variable in the state variable; the control variable is defined as: ; wherein, represents a control variable, is an inertia matrix in the dynamic model, is a joint torque, is the disturbance torque, is a preset nonlinear resultant force term; the output variable is defined as: , represents an output variable, represents an output matrix , represents a state vector of the robot limb system.

[0009] In an implementation form of the present application, based on a state feedback optimal control principle, the dynamic model and an output feedback controller constructed based on the output variable, a first robust control equation is determined, specifically comprising: determining a state feedback controller based on the state feedback optimal control principle and the dynamic model ; wherein, , is a state feedback gain matrix, represents a control strategy weight matrix, is a second coefficient matrix , is a positive definite gain matrix for solving Riccati equation; constructing a corresponding state optimal performance index function according to the state feedback controller; the state optimal performance index function comprises the preset state weight matrix; reconstructing the output feedback controller constructed based on the output variable, the state optimal performance index function and the dynamic model to obtain the first robust control equation; the output feedback controller comprises the output variable and the control gain matrix.

[0010] In an implementation form of the present application, the first robust control equation is: ; wherein, , is a first coefficient matrix , is the control gain matrix, represents an external input torque, represents the preset state weight matrix.

[0011] In an implementation form of the present application, based on a mapping relationship between the state variable and the output variable in the dynamic model, the first robust control equation is converted into a second robust control equation, specifically comprising: transforming a state vector of the first robust control equation according to the dynamic model to obtain: ; according to the mapping relationship between the state variable and the output variable , a second robust control equation is obtained: .

[0012] In an implementation form of the application, the to-be-estimated parameter vector is determined according to the second robust control equation, specifically comprising: The second robust control equation is subjected to Kronecker product deformation to extract the to-be-estimated parameter vector according to the deformed parameter expression; the to-be-estimated parameter vector is established by vectorization operation based on the positive definite gain matrix; The online estimation is performed based on a preset adaptive law to obtain the control gain matrix, specifically comprising: The online estimation is performed based on the preset adaptive law The online estimation is performed to determine an online estimation parameter vector of the to-be-estimated parameter vector when an estimation error of the to-be-estimated parameter vector is less than a predetermined value; ; wherein, is a time derivative of the online estimation parameter ; is a preset learning gain; is an auxiliary vector; The positive definite gain matrix is determined according to the determined online estimation parameter and the to-be-estimated parameter vector, so as to determine the control gain matrix according to the positive definite gain matrix and the second robust control equation.

[0013] In an implementation form of the application, wherein, is a predefined auxiliary regression matrix; represents an estimation error, ; is a true value parameter vector corresponding to the to-be-estimated parameter vector.

[0014] In a second aspect, the embodiments of the application further provide an output feedback robust control system for a humanoid robot, which can execute the output feedback robust control method for a humanoid robot described above; the system comprises: An establishing module is configured to establish a dynamics model of a robot limb end system and define state variables, control variables and output variables; wherein the dynamics model is constructed based on Lagrange equation and at least includes an inertia matrix, a Coriolis force vector, a gravity vector and a disturbance torque; the state variables are motion state parameters of limb end joints, the control variables are control input parameters acting on the limb end joints, and the output variables are measurable limb end joint state parameters; A determining module is configured to determine a first robust control equation based on a state feedback optimal control principle, the dynamics model and an output feedback controller constructed with the output variables; wherein the first robust control equation is added with a preset state weight matrix and a control strategy weight matrix to adjust control performance. a conversion module, configured to convert the first robust control equation into a second robust control equation based on a mapping relationship between the state variable and the output variable in the dynamic model, to determine a to-be-estimated parameter vector according to the second robust control equation, and to obtain a control gain matrix based on a preset adaptive law and online estimation; a generation module, configured to generate a corresponding control variable based on the control gain matrix and the output variable acquired in real time, and to apply the control variable to a robot limb end joint to perform robust control on the robot limb end system.

[0015] In a third aspect, the embodiments of the present application further provide an output feedback robust control device for a humanoid robot, the device comprising: at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the output feedback robust control method for a humanoid robot described above.

[0016] Compared with the prior art, the present application has the following remarkable effects: Through the above technical solution, the present application proposes an output feedback robust control scheme based on data driving to realize optimal robust control of a humanoid robot system. The present application eliminates the dependence of robot limb end joint control on state variables by reconstructing a robust control equation, and specifically overcomes the problem of difficult state observation of a humanoid robot. Secondly, the dependence of robot control on internal states is eliminated through equation transformation, and control can be realized only through measurable joint state parameters. Moreover, an adaptive law is designed to realize real-time update of a control gain matrix, so that efficient and dynamic stable control can be realized under the condition of less data. Furthermore, an output feedback, robust control and data driving integrated humanoid robot limb end robust control scheme is constructed, which overcomes the limitations of model dependence, disturbance resistance capability and control efficiency, and can perform precise output feedback robust control for a system with complex dynamic characteristics of a humanoid robot. BRIEF DESCRIPTION OF DRAWINGS

[0017] The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application and illustrate the illustrative embodiments of the present application and its description, and do not constitute improper limitations on the present application. In the drawings: Figure 1 FIG. 1 is a flowchart of an output feedback robust control method for a humanoid robot according to an embodiment of the present application; Figure 2A lower limb structure diagram of a humanoid robot is applied to a humanoid robot output feedback robust control method in the embodiments of the present application. Figure 3 A humanoid robot robust control system flow chart is provided in the embodiments of the present application. Figure 4 A robot limb end system stable state simulation curve diagram after the output feedback robust control method is used in the embodiments of the present application. Figure 5 A structure diagram of a humanoid robot output feedback robust control system is provided in the embodiments of the present application. Figure 6 A structure diagram of a humanoid robot output feedback robust control device is provided in the embodiments of the present application. DETAILED DESCRIPTION

[0018] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described clearly and completely below in combination with the embodiments of the present application and corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0019] The existing humanoid robot system is a complex system with multiple joints, strong coupling and high nonlinearity. Using a model-dependent control method often requires a large number of system parameters, and using a model-free control method with poor interpretability has the problem of a long training process. Common control algorithms, such as PID control, adaptive control, model predictive control, etc., the PID control algorithm has low modeling requirements for the system and is widely used in industrial robot systems. However, for humanoid robots, the effectiveness of traditional control strategies represented by PID is limited, and the anti-interference ability is weak in the face of sudden external force collision, road disturbance and other scenes, and the system state recovers slowly after being disturbed.

[0020] Based on this, the embodiments of the present application provide a humanoid robot output feedback robust control method, system and device to solve the technical problem of how to overcome the limitations of model dependence, anti-interference ability, control efficiency, etc. for the complex dynamic characteristics of the humanoid robot system, and to perform precise output feedback robust control.

[0021] The various embodiments of the present application will be described in detail below in combination with the drawings.

[0022] The embodiments of the present application provide a humanoid robot output feedback robust control method, as shown in Figure 1 The method can include steps S101-S104: S101, a dynamics model of a robot limb system is established, and state variables, control variables and output variables are defined.

[0023] The dynamics model is constructed based on Lagrange equation, and at least includes an inertia matrix, a Coriolis force vector, a gravity vector and a disturbance torque. The state variables are motion state parameters of limb joints, the control variables are control input parameters acting on the limb joints, and the output variables are measurable limb joint state parameters.

[0024] In the embodiments of the present application, the above method is specifically applied to output feedback robust control of a lower limb of a humanoid robot. The robot limb system includes a plurality of lower limb joints with one or more degrees of freedom. The lower limb joints at least include a hip joint, a thigh joint, a leg joint, a knee joint, an ankle joint and a waist joint. As shown in the figure, the hip joint (front and back) can be understood as the hip joint having two degrees of freedom of front and back, and the hip joint (everting) refers to the degree of freedom of the hip joint turning outward. The robot lower limb of the present application is set to five joints of a single leg, specifically a hip joint, a thigh joint, a leg joint, a knee joint and an ankle joint. One joint is set in the waist. This setting can ensure the stability of the realized robot and its walking and other movements. Figure 2

[0025] The present application establishes a dynamics model of a robot limb system, specifically including: Based on the pose vectors corresponding to each lower limb joint, the kinematic equation of the humanoid robot lower limb is modeled to obtain a motion equation. Based on the motion equation and the Lagrange equation, a dynamics model is established.

[0026] The pose vector is specifically as shown in the figure, , which can be understood as the position of each driving joint of the lower limb, is the attitude of the humanoid robot in the world coordinate system. Then, the kinematic equation of the humanoid robot is modeled, and since the motion of the humanoid robot is also caused by the mixed action of its own active force and the environmental passive force, the following motion equation is obtained, as shown in formula (1): (1) wherein, , , represents an inertia matrix corresponding to the position and attitude in the robot limb system, is the transpose matrix of ; and are acceleration vectors corresponding to the attitude and joint position, respectively; and are first and second bias forces caused by centrifugal force, Coriolis force and gravity, respectively,​ joint driving torque corresponding to a certain joint, , generalized contact force of the external environment corresponding to the posture and joint position respectively.

[0027] Subsequently, based on the Lagrange dynamics equation, the humanoid robot is converted into a multi-link dynamic system, and the kinetic energy and potential energy of the system can be used to establish the Lagrange equation, as shown in equation (2): (2) wherein, represents the inertia matrix of the system, represents the gravity matrix of the system, is the robot limb system selection matrix, is the robot joint torque, is the Jacobian matrix, is the external input torque, is defined as the system parameter in a broad sense.

[0028] The general dynamics model of the lower limbs of the humanoid robot can be obtained by simplifying the Lagrange dynamics equation: (3) wherein, are the position, velocity, and acceleration vectors of the humanoid robot, is the inertia matrix of the humanoid robot, is the Coriolis force vector of the humanoid robot, is the gravity vector, is the control input torque of the robot, is the disturbance torque.

[0029] At the same time, for the convenience of calculation, the above dynamics model can also be defined as: (4) wherein, is is a nonlinear resultant force term. In actual scenarios, the humanoid robot has a certain load capacity, and the load will cause the center of mass of the humanoid robot to change due to the difference in load. This is similar to an external force disturbance given to the robot system, both of which affect the center of mass of the robot, thereby causing the uncertainty of the vector matrices and , thereby making the model uncertain. Therefore, the following realistic conditions are used to solve the problem of load change. First, there exist positive definite matrices and such that: second, it is assumed that there exist vectors and a non-negative function i.e. .

[0030] Therefore, for the robot dynamic model, the state variable is defined as formula (5): (5) wherein, , represents the position coordinates of the lower limb joint, represents the posture of the lower limb of the humanoid robot in the world coordinate system, represents the pose vector of the limb end joint, is the joint velocity vector, is the first variable in the state variable, is the second variable in the state variable.

[0031] The control variable is defined as formula (6): (6) wherein, represents the control variable, is the inertia matrix in the dynamic model, is the joint torque, is the disturbance torque, is the preset nonlinear resultant force term.

[0032] From the state equation, we can get , and then use the formula of the robot system dynamics model to derive the state space expression, as shown in formula (7): (7) Therefore, the nonlinear variable is obtained, as shown in formula (8): (8) Further, the nonlinear control model of the humanoid robot limb end system is obtained, as shown in formula (9): (9) wherein, is the first coefficient matrix , is the second coefficient matrix , represents the output matrix , , represents the output variable, that is, the position of each lower limb joint of the humanoid robot, that is, the measurable limb end joint state parameter. Further, the output variable can be defined as: .

[0033] It should be noted that the execution subject of the output feedback robust control method for the humanoid robot described above can be a special computing unit arranged inside the humanoid robot, for example, the special computing unit can include a Field Programmable Gate Array (FPGA) chip, a Graphics Processing Unit (GPU), a Microcontroller Unit (MCU), etc., and the types listed above are only exemplary, and the execution subject is not limited thereto, and the present application does not make a specific limitation thereon.

[0034] In the embodiment of the present application, the first robust control equation is determined based on the state feedback optimal control principle, the dynamic model and the output feedback controller constructed based on the output variable, and specifically includes:

[0035] The first robust control equation is added with a preset state weight matrix and a control strategy weight matrix to adjust the control performance.

[0036] In the embodiment of the present application, the first robust control equation is determined based on the state feedback optimal control principle, the dynamic model and the output feedback controller constructed based on the output variable, and specifically includes: The state feedback controller is determined according to the dynamic model based on the state feedback optimal control principle . Wherein, , is a state feedback gain matrix, represents a control strategy weight matrix, is a second coefficient matrix , is a positive definite gain matrix for solving Riccati equation. According to the state feedback controller, a corresponding state optimal performance index function is constructed. The state optimal performance index function includes a preset state weight matrix. The output feedback controller constructed based on the output variable, the state optimal performance index function and the dynamic model are reconstructed simultaneously to obtain the first robust control equation. The output feedback controller includes an output variable and a control gain matrix.

[0037] Specifically, the nominal system of the robot limb end system control function is obtained based on the above formula (9), as shown in formula (10): (10) Further, according to the state feedback optimal control principle, formula (11) is obtained: (11) Wherein, is a control strategy weight matrix of the humanoid robot system, R represents an m-dimensional vector space in a real number field, which can adjust the weight of each part in the state feedback controller.

[0038] The Riccati equation is solved by the positive definite gain matrix P, as shown in equation (12): (12) The state optimal performance index function is shown in equation (13): (13) Wherein, represents the value of the state optimal performance index function.

[0039] At this point, the optimal performance index function of the state feedback controller is established.

[0040] Subsequently, the output feedback optimal performance index function is designed for the humanoid robot system equation as shown in equation (14): (14) Wherein, is the value of the output feedback optimal performance index function, in order to solve the state and the control input of the system may not be 1:1 corresponding relationship in the actual system, therefore, the weight matrix is introduced to adjust the system, that is is the preset state weight matrix of the humanoid robot system, which can be set by expert experience, and is not limited here, represents an n-dimensional vector space in a real number field. The index function is divided into three parts, the first part is for processing system uncertainty disturbance, the second part determines the convergence rate of system state, and the third part determines the optimization of system control. The preset state weight matrix Q can reasonably allocate the proportion of each state, and the weight influence rate in the performance index function is positive feedback.

[0041] In order to use the system output scalar y as the output feedback control variable, the robot output feedback controller is set as follows equation (15): (15) Wherein, is the control gain matrix to be solved, which is a constant matrix, and here * is to distinguish from the above state feedback control variable u. In fact, both of them are the same control variable set, that is, they are equivalent in the control process.

[0042] Then, the humanoid robot system robust control equation is reconstructed by combining the robot output feedback controller Substitute equation (13) into equation (16): (16) Take the derivative of the system time to get equation (17): (17) Substituting the above humanoid robot nominal system, formula (18) is obtained: (18) Further, the first robust control equation of the humanoid robot is obtained, as formula (19): (19) Wherein, .

[0043] S103, based on the mapping relationship between the state variables and the output variables in the dynamic model, the first robust control equation is converted into the second robust control equation, so as to determine the to-be-estimated parameter vector according to the second robust control equation, and the control gain matrix in the output feedback controller is obtained based on the preset adaptive law. The second robust control equation is constructed based on the output variables.

[0044] In the embodiment of the present application, based on the mapping relationship between the state variables and the output variables in the dynamic model, the first robust control equation is converted into the second robust control equation, specifically including: According to the dynamic model, the state vector of the first robust control equation is transformed to obtain: According to the mapping relationship between the state variables and the output variables , the second robust control equation is obtained: .

[0045] Specifically, for the robust control equation of the constructed humanoid robot, the intention is to obtain the control solution by using the system output variable y, so it is necessary to transform the humanoid robot system state x, as formula (20): (20) Then, according to the mapping relationship between the state variables and the output variables , the second robust control equation is obtained, as shown in formula (21): (21) Further calculation can obtain formula (22): (22) At this point, by using parameter transformation, the influence of the system state variable x is removed, and accurate output feedback robust control can be realized only by using the system output variable y without the state variable x. The dependence on the internal state is completely removed, and a model that can realize control only by measurable joint position y is constructed.

[0046] Furthermore, the humanoid robot end effector system of this application is a continuous-time system, and the obtained control function is also affected by time. Therefore, a single solution cannot obtain the solution of the system control function. Thus, the humanoid robot control function obtained in this application adopts an online solution method. During the solution process, it is necessary to determine the parameter vector to be estimated based on the second robust control equation, specifically including: The second robust governing equation is transformed using the Kronecker product to extract the estimated parameter vector from the transformed parameter expression. The estimated parameter vector is established based on the positive definite gain matrix through vectorization operations.

[0047] In other words, this application uses the Kronecker product to obtain a parameterized expression, as shown in formula (23): (twenty three) To make the formula more concise, the variable is defined as formula (24): (twenty four) This completes the aforementioned positive definite gain matrix of variables. to the parameter vector to be estimated The online transformation is used to extract the parameter vector to be estimated. , The calculation utilizes adaptive law online estimation, by The estimated value Obtain the unknown matrix .

[0048] In one embodiment of this application, the above-mentioned online estimation based on a preset adaptive law to obtain the control gain matrix specifically includes: By pre-setting an adaptive law Perform online estimation to determine the online estimation parameter vector. estimation error If the value is less than the predetermined value, determine the online estimated parameters corresponding to the parameter vector to be estimated. .in, For online parameter estimation The time derivative. This is the preset learning gain. This is an auxiliary vector. Based on the determined online estimation parameters... Given the parameter vector to be estimated, determine the positive definite gain matrix, and then determine the control gain matrix based on the positive definite gain matrix and the second robust control equation.

[0049] The online estimation method is as follows: Predefined vectors With auxiliary regression matrix For example, in formula (25): (25) In the present application, the following notations are used: denotes a value taken from a certain (arbitrary numerical) range in the real number field, which is explained below. Design parameters , define the auxiliary vector , as shown in equation (26): (26) wherein denotes the estimation error, , is the true value parameter vector corresponding to the to-be-estimated parameter vector. It can be understood as the difference between the estimated weight matrix and the actual weight matrix. Substituting , the auxiliary vector can be derived.

[0050] The adaptive law for estimating the unknown parameter can be designed as equation (27): (27) wherein the learning gain . When the parameter is estimated online using the adaptive law, if the regression vector satisfies the persistent excitation condition, the estimation error of the estimated parameter exponentially converges to zero, i.e., the estimated parameter exponentially converges to the true value . The above predetermined value is 0, and in actual use, the specific value is set by the user according to the scene or expert experience, which is not limited here.

[0051] Based on this, the solution of the above can be obtained, and then the matrix is obtained according to . Further, the control gain matrix is obtained by solving the second robust control equation.

[0052] Wherein, the specific process of the online estimation algorithm is as follows: 1) Initial condition: set the initial condition and the learning parameter of the adaptive law, ; 2) Measure parameters: measure the system input and the augmented system output (the augmented system output is the actual value of y in the above equation (22), which is the change value generated by the robot system, which can be obtained by external measurement without solving equations), and then perform P-W transformation using equation transformation and matrix augmentation; 3) Online solution: calculate , , , and online update unknown parameters , obtain control gain matrix ; 4) Control application: apply output feedback control to the augmented system .

[0053] S104, based on the control gain matrix and the real-time acquired output variable, a corresponding control variable is generated to be applied to the robot limb end joint to perform robust control on the robot limb end system.

[0054] In the embodiment of the application, after the control gain matrix is constructed by S101-S103, the real-time acquired output variable can be input into the output feedback controller, and then the control variable based only on the output variable can be obtained, and then the robust control of the robot limb end joint is performed through the control variable.

[0055] In the embodiment of the application, Figure 3 is a humanoid robot robust control system flowchart, including establishing a humanoid robot model, deriving a robot lower limb control model (i.e. a dynamics model), establishing a robust control system equation (establishing a first robust control equation), then constructing an optimal control performance index function (i.e. constructing a state optimal performance index function), then data-driven learning (i.e. the process of converting the first robust control equation to the second robust control equation), and then reconstructing the control equation (obtaining the second robust control equation); next, the output of the controlled humanoid robot system is performed.

[0056] Based on the above scheme, the stable state simulation curve of the robot limb system obtained by applying artificial external interference to the humanoid robot system is as shown in Figure 4 . As shown in Figure 4 , the curves of different colors are the state simulation curves of the robot recovering to a stable state after applying artificial external interference of different intensities. The initial state parameters of the robot are not stable, that is, the humanoid robot system is subjected to artificial external interference, and the state of the robot lower limb is abnormal, that is, in a non-stable state. The system is controlled by using the above output feedback robust control method, and the state curve of the system converges quickly, and the system gradually sets to a stable behavior, highlighting the ability of the proposed humanoid robot output feedback robust control method to stabilize the system, that is, the stable control of the humanoid robot system is realized only with the system output variable y. The simulation results show the effectiveness of the output feedback optimal robust control for the control of the robot end limb.

[0057] By the technical solution, the application proposes an output feedback robust control scheme based on data driving to realize optimal robust control of the humanoid robot system. The application eliminates the dependence of the robot limb end joint control on the state variable by reconstructing the robust control equation, and specifically overcomes the problem of difficult state observation of the humanoid robot. Secondly, the dependence of the robot control on the internal state is eliminated through equation transformation, and the control can be realized only through the measurable joint state parameters. Moreover, the adaptive law is designed to realize real-time update of the control gain matrix, so that efficient and dynamic stable control can be realized under the condition of less data. Further, the humanoid robot limb end robust control scheme integrating output feedback, robust control and data driving is constructed, which overcomes the limitations of model dependence, disturbance resistance ability and control efficiency, and can realize precise output feedback robust control for the system with complex dynamic characteristics of the humanoid robot.

[0058] In addition, the application applies the output feedback robust optimal control idea to the robot system control algorithm, that is, for the stabilization control of the lower limbs of the robot. When the humanoid robot is under abnormal physical disturbance, the lower limbs may not stand stably. The control method of the application can effectively control the joint angle of the humanoid robot when the posture of the humanoid robot is abnormal, so as to avoid loss caused by instability and loss of control of the humanoid robot. The application of the output feedback robust control method effectively improves the control precision of the lower limbs of the humanoid robot, and avoids the requirement for the internal state of the system in the control algorithm of the humanoid robot. As a kind of control applied to the precision industrial robot, the application of the output feedback robust control method to the control of the humanoid robot system is an innovative work, which can fully exert the advantages of optimal control and significantly improve the control effect.

[0059] Figure 5 A structural schematic diagram of an output feedback robust control system 500 for a humanoid robot provided by the embodiment of the application is shown in FIG. 5. Figure 5 As shown in FIG. 5, the output feedback robust control system 400 for the humanoid robot adopts the output feedback robust control method for the humanoid robot. The output feedback robust control system 500 for the humanoid robot includes: The establishing module 501 is configured to establish a dynamics model of the robot limb tip system, and define state variables, control variables and output variables. The dynamics model is constructed based on Lagrange equation, and at least includes an inertia matrix, a Coriolis force vector, a gravity vector and a disturbance torque. The state variables are motion state parameters of the limb tip joints, the control variables are control input parameters acting on the limb tip joints, and the output variables are measurable limb tip joint state parameters. The determining module 502 is configured to determine a first robust control equation based on a state feedback optimal control principle, the dynamics model and an output feedback controller constructed based on the output variables. The first robust control equation is added with a preset state weight matrix and a control strategy weight matrix to adjust the control performance. The converting module 503 is configured to convert the first robust control equation into a second robust control equation based on a mapping relationship between the state variables and the output variables in the dynamics model, so as to determine a to-be-estimated parameter vector according to the second robust control equation, and perform online estimation based on a preset adaptive law to obtain a control gain matrix. The second robust control equation is constructed based on the output variables. The generating module 504 is configured to generate corresponding control variables based on the control gain matrix and the output variables acquired in real time, and apply the control variables to the robot limb tip joints, so as to perform robust control on the robot limb tip system.

[0060] Figure 6 A structural schematic diagram of an output feedback robust control device for a humanoid robot provided by an embodiment of the present application is shown in FIG. 1. As shown in FIG. 1, the device includes: Figure 6 at least one processor; and a memory connected with the at least one processor in communication. The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: ​A dynamics model of the robot limb system is established, and state variables, control variables and output variables are defined. The dynamics model is constructed based on Lagrange equation, and at least includes an inertia matrix, a Coriolis force vector, a gravity vector and a disturbance torque. The state variables are motion state parameters of the limb joints, the control variables are control input parameters acting on the limb joints, and the output variables are measurable limb joint state parameters. Based on the state feedback optimal control principle, the dynamics model and an output feedback controller constructed based on the output variables, a first robust control equation is determined. The first robust control equation is added with a preset state weight matrix and a control strategy weight matrix to adjust the control performance. Based on the mapping relationship between the state variables and the output variables in the dynamics model, the first robust control equation is converted into a second robust control equation, so as to determine a to-be-estimated parameter vector according to the second robust control equation, and to obtain a control gain matrix in the output feedback controller based on a preset adaptive law for online estimation. The second robust control equation is constructed based on the output variables. Based on the control gain matrix and the real-time acquired output variables, corresponding control variables are generated to be applied to the robot limb joints, so as to robustly control the robot limb system.

[0061] Each of the embodiments in the present application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the difference from other embodiments. In particular, for the system and device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the related parts can be referred to the part of the method embodiments.

[0062] The system and device provided by the embodiments of the present application correspond to the method, and therefore, the system and device also have similar beneficial technical effects as the method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the system and device will not be described here.

[0063] It should also be noted that the terms “comprising”, “containing”, or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such a process, method, article or device. Without more limitations, the element defined by the statement “comprising a” does not exclude the presence of additional identical elements in the process, method, article or device including the element.

[0064] The above only describes the embodiments of the present application and is not intended to limit the present application. The present application can have various changes and modifications for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the scope of the claims of the present application.

Claims

1. An output feedback robust control method for a humanoid robot, characterized by, The method comprises: establishing a dynamics model of a robot limb system, and defining state variables, control variables and output variables; wherein the dynamics model is constructed based on Lagrange equation, and at least includes an inertia matrix, a Coriolis force vector, a gravity vector and a disturbance torque; the state variables are motion state parameters of limb joints, the control variables are control input parameters acting on the limb joints, and the output variables are measurable limb joint state parameters; determining a first robust control equation based on a state feedback optimal control principle, the dynamics model and an output feedback controller constructed based on the output variables; wherein the first robust control equation is added with a preset state weight matrix and a control strategy weight matrix to adjust the control performance; based on a mapping relationship between the state variables and the output variables in the dynamics model, converting the first robust control equation into a second robust control equation to determine a to-be-estimated parameter vector according to the second robust control equation, and performing online estimation based on a preset adaptive law to obtain a control gain matrix in the output feedback controller; the second robust control equation is constructed based on the output variables; based on the control gain matrix and the output variables acquired in real time, generating corresponding control variables to be applied to the robot limb joints to perform robust control on the robot limb system.

2. The output feedback robust control method for a humanoid robot according to claim 1, wherein The method is applied to output feedback robust control of a lower limb of a humanoid robot; the robot limb system comprises a plurality of lower limb joints with one or more degrees of freedom; the lower limb joints at least include a hip joint, a thigh joint, a knee joint, an ankle joint and a waist joint; establishing a dynamics model of a robot limb system, specifically comprising: based on a pose vector corresponding to each of the lower limb joints, modeling a kinematics equation of the lower limb of the humanoid robot to obtain a motion equation; based on the motion equation and the Lagrange equation, establishing the dynamics model.

3. The output feedback robust control method for a humanoid robot according to claim 1, wherein the state variables are defined as: ; wherein , represents a position coordinate of a lower limb joint, represents a posture of a lower limb of a humanoid robot in a world coordinate system, represents a pose vector of a distal joint, is a joint velocity vector, is a first variable in a state variable, is a second variable in a state variable; the control variables are defined as: ; wherein denotes a control variable, is an inertia matrix in the dynamics model, is a joint torque, is a disturbance torque, is a preset nonlinear resultant force term; The output variable is defined as: , The output variable is defined as: The output matrix is defined as: , The state vector of the robot limb system is defined as:

4. The output feedback robust control method for a humanoid robot according to claim 3, wherein based on a state feedback optimal control principle, the dynamics model and an output feedback controller constructed based on the output variables, determining a first robust control equation, specifically comprising: determining a state feedback controller from the dynamic model based on the state feedback optimal control principle ; wherein , is a state feedback gain matrix, denotes a control policy weight matrix, is a second coefficient matrix , is a positive definite gain matrix for solving Riccati equation; constructing a corresponding state optimal performance index function according to the state feedback controller; the state optimal performance index function includes the preset state weight matrix; reconstructing the output feedback controller constructed based on the output variables, the state optimal performance index function and the dynamics model to obtain the first robust control equation; the output feedback controller includes the output variables and the control gain matrix.

5. The output feedback robust control method for a humanoid robot according to claim 4, wherein the first robust control equation is: ; wherein , is a first coefficient matrix , is the control gain matrix, denotes an external input torque, denotes the preset state weight matrix.

6. The output feedback robust control method for a humanoid robot according to claim 5, wherein based on a mapping relationship between the state variables and the output variables in the dynamics model, converting the first robust control equation into a second robust control equation, specifically comprising: According to the kinetic model, a transformation is performed on the state vector of the first robust control equation to obtain: ; According to the mapping relationship between the state variable and the output variable , a second robust control equation is obtained: .

7. The output feedback robust control method for a humanoid robot according to claim 6, wherein determining a to-be-estimated parameter vector according to the second robust control equation, specifically comprising: performing Kronecker product deformation on the second robust control equation to extract the to-be-estimated parameter vector according to the parameter expression after deformation; the to-be-estimated parameter vector is established based on the positive definite gain matrix through vectorization operation; The control gain matrix is obtained through online estimation based on a preset adaptive law, and specifically includes: by the preset adaptive law performing online estimation to determine an estimation error of an online estimation parameter vector when the estimation error is less than a predetermined value, determining the online estimation parameter corresponding to the to-be-estimated parameter vector ; wherein, is a time derivative of the online estimation parameter ; is a preset learning gain; is an auxiliary vector; determining the positive definite gain matrix according to the determined online estimation parameters and the parameter vector to be estimated, to determine the control gain matrix according to the positive definite gain matrix and the second robust control equation.

8. The output feedback robust control method for a humanoid robot according to claim 7, wherein wherein, is a predefined auxiliary regression matrix; denotes an estimation error, ; is the true parameter vector corresponding to the parameter vector to be estimated.

9. An output feedback robust control system for a humanoid robot, characterized by, The system can perform the output feedback robust control method for a humanoid robot according to any one of claims 1-8; the system comprises: The system comprises: The system comprises: The system comprises: The system comprises:

10. An output feedback robust control device for a humanoid robot, characterized by, The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises: The system comprises

Images