Design method of adaptive robust control wheel-legged robot based on cooperative game theory

Through the adaptive and robust control method based on cooperative game theory, the problem of insufficient motion control accuracy of the two-foot wheel leg robot in complex environments is solved, efficient and stable motion and environmental adaptability are achieved, and the accuracy of the task is ensured and efficiency.

CN119002269AInactive Publication Date: 2024-11-22ANHUI UNIV

Patent Information

Application Number
CN202411074038.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-07
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing control algorithms are difficult to effectively deal with changes in system internal parameters and external perturbations in complex terrain and dynamic environments, resulting in insufficient motion control accuracy of bifoot wheel leg robots, especially when facing uncertainty and nonlinear constraints.

Method used

Adaptive robust control method based on cooperative game theory is adopted, and the adaptive robust control algorithm is designed by establishing a dynamic model of fuzzy underdrive system with uncertain parameters, and the control parameters are optimized through the 'Pareto optimal' method to balance Performance indicators ensure efficient motion and stability of the system in complex environments.

Benefits of technology

It significantly improves the robot's trajectory tracking accuracy and stability in complex terrain, enhances environmental adaptability and robustness, ensures the efficiency and accuracy of task execution, and reduces the risk of control failure caused by uncertainty in the initial condition.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119002269A_ABST
    Figure CN119002269A_ABST
Patent Text Reader

Abstract

The invention provides an adaptive robust control wheel-legged robot design method based on a cooperative game theory, and the method comprises the following steps: S1, building a fuzzy underactuated system (UMS) dynamic model with uncertain parameters, and writing a dynamic model of a jumping process of a biped wheel-legged robot; s2, in the control process, a specified track is expressed as a servo constraint applied to the fuzzy UMS; s3, designing an adaptive robust control algorithm by considering the uncertainty of the wheel-legged robot system; s4, constructing a performance index in a control process based on a cooperative game theory; and S5, performing numerical simulation, observing a result, analyzing a control effect, and finally giving a conclusion. According to the method, the dynamic model with the uncertain parameters is constructed by introducing the fuzzy set theory, the uncertainty that the internal parameters of the system change along with time and are limited can be effectively dealt with, and the trajectory tracking precision of the robot in the complex terrain and the dynamic environment is remarkably improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical system dynamics control, and in particular to a design method of an adaptive robust control wheel-legged robot based on cooperative game theory. Background Art

[0002] In the wave of technological innovation, legged robots and wheeled robots are gradually becoming the key force to promote social progress. With its excellent terrain adaptability, legged robots have shown great potential in search and rescue, field exploration and other fields. They can replace humans to enter dangerous or difficult-to-reach areas, perform key tasks and ensure the safety of personnel. Wheeled robots, with their efficient and flexible operating performance, provide support in logistics distribution, home services, medical assistance and other aspects. They improve efficiency through automation and reduce the daily burden of humans. The design and manufacture of these two types of robots must take into account both practicality and economy. They must not only meet basic functions, but also pursue ease of operation, compact structure and controllable costs. They are not just tools, but also important media for improving the quality of life and promoting social inclusion. The development of legged and wheeled robots is not only about scientific and technological progress, but also about how to bring more convenient, safe and rich life experience to all mankind through technological innovation. Therefore, designing robots that are both economical and comprehensive has become the focus of common concern of the scientific and technological community and all sectors of society.

[0003] The research and development of wheel-legged robots is committed to combining the dual advantages of foot-type and wheel-type robots, aiming to improve their adaptability and movement efficiency in changing terrain. This type of robot is particularly good at maintaining stability on uneven ground, while being able to quickly cross flat areas, making it have broad application prospects in emergency rescue, complex environment exploration, and special industrial operations. The excellent posture control ability of wheel-legged robots makes it more accurate and reliable when performing delicate operations, and the combination with visual radar and robotic arms further expands its functions, enabling it to perform complex tasks such as object recognition, grasping and handling. This design not only enhances the autonomy of the robot, but also greatly enhances its application potential in the fields of automation and intelligence, indicating that future robotics technology will bring more convenience and security to society.

[0004] In complex terrain and dynamic environments, if bipedal wheeled leg robots use traditional control algorithms such as PID (proportional-integral-differential) control and traditional sliding mode control (SMC), they may face challenges in precise motion control. The limitations of these algorithms in terms of robustness are mainly reflected in their lack of adaptability to changes in internal system parameters or external disturbances. For example, although PID control is simple and works well in some applications, it lacks robustness to complex dynamics and uncertain factors. Although traditional SMC is effective for some nonlinear systems, it may be difficult to maintain high performance in the face of parameter uncertainty and external disturbances, and its control signal may contain high-frequency jitter, affecting the stability of the system. In addition, these control algorithms may not be ideal in terms of constraint processing, especially when dealing with non-holonomic constraints and nonlinear constraints, it may not be possible to ensure that the robot's motion meets all operating restrictions, resulting in insufficient control accuracy. Therefore, in order to improve the adaptability and control accuracy of bipedal wheeled leg robots in changing environments, it is necessary to develop more advanced control algorithms to better handle uncertainties and system constraints. Summary of the invention

[0005] The purpose of the present invention is to provide a design method for an adaptive robustly controlled wheel-legged robot based on cooperative game theory, aiming to solve the problems raised by the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: a design method of an adaptive robust control wheel-legged robot based on cooperative game theory, comprising the following steps:

[0007] S1. Establish a dynamic model of the fuzzy underactuated system with uncertain parameters, and then write a dynamic model of the jumping process of the bipedal wheel-legged robot;

[0008] S2, during the control process, the specified trajectory is expressed as a servo constraint imposed on the fuzzy UMS;

[0009] S3. Design of adaptive robust control algorithm considering the uncertainty of wheel-legged robot system;

[0010] S4, based on cooperative game theory to build the performance indicators of the control process, and then use the "Pareto optimal" method to find the optimal parameters that can balance the two performance indicators, which are the optimal values ​​of the two important parameters needed to control a complex system;

[0011] S5. Conduct numerical simulation, observe the results, analyze the control effect, and finally draw conclusions.

[0012] The form of establishing the dynamic model of a bipedal wheeled leg robot system with uncertain parameters specifically includes: considering a dynamic model of a bipedal wheeled leg robot system with uncertain parameters, a control dynamic model of a fuzzy system is described as follows:

[0013]

[0014] Here, t∈R n is time, χ∈R n are generalized coordinates, It's speed. is the acceleration, χ, and Respectively represent the position, velocity and acceleration of the joints in the bipedal wheeled leg robot system. p , σ represents the uncertainty parameter in the mechanical system, which may change over time. The symbol "Σ" is used to define the accuracy range of "σ". Although there is uncertainty, its range of variation is limited. "τ" is used to represent the control input to the system, B∈R n×m is the coefficient matrix of the control input. M, C and G are all continuous, and M(χ,σ,t), including parameters related to inertia, represents the inertia matrix of the system. and G(χ,σ,t) represent Coriolis / centrifugal force and gravity respectively, and F(χ,σ,t) is friction force. Friction effect, as a common nonlinear natural phenomenon, is widely present in various mechanical systems. In order to reduce the adverse effects of friction on the control performance of mechanical systems, many researchers have proposed a variety of friction modeling methods to deeply explore and grasp the essence of this phenomenon. These models include Coulomb friction model, viscous friction model, Stribeck friction model and Dalbe dynamic friction model, each of which is committed to revealing the core mechanism of friction control.

[0015] The dynamic model of the jumping process of the bipedal wheeled robot can be described as follows:

[0016] The kinetic energy of the body is:

[0017]

[0018] The potential energy of the body is:

[0019] V=Mgy (3)

[0020] Lagrangian operator:

[0021]

[0022] Select the joint motor angle α as the generalized coordinate, the generalized force as the torque τ of the four joint motors, and list the Lagrange equation:

[0023]

[0024] Expanding and simplifying formula (5) yields:

[0025]

[0026] The expressions of y1 and y2 are as follows:

[0027]

[0028] During the control process, the specified trajectory is expressed as a servo constraint imposed on the fuzzy UMS. First, all uncertainties in Equation (1) are bounded, and their boundaries are described by known fuzzy sets, which can be expressed as:

[0029]

[0030] where σ i is the i-th term of σ, is the fuzzy set of σ i , ∑i is the universe of discourse of σ i , and it is compact. μ i : σ i →[0,1] is the membership function of . The tracking control force τ is provided by the actuator, which makes the fuzzy UMS (1) follow the specified tracking trajectory. At this time, the tracking control problem is considered in the framework of constrained motion, and the specified tracking trajectory is regarded as a servo constraint imposed on the fuzzy UMSs.

[0031] Assume that the fuzzy UMS is subject to holonomic servo constraints in the form of:

[0032] q i (χ,t) = 0, i = 1,2,...,h (9)

[0033] and m - h non - holonomic servo constraints in the form of:

[0034]

[0035] where m < n, and we assume that these constraint equations are differentiable. Combine the first - order derivative of (9) with respect to t and (10), and express the final set of equations in matrix form as:

[0036]

[0037] where A = [A li m×n , c = [c1, c2,..., c m . Then, combine the second - order derivative of (9) with respect to t and the first - order derivative of (10) with respect to t, and express the result in matrix form as:

[0038]

[0039] ​Where b=[b1,b2,…,b m ]. Here equation (11) controls the speed (12) Controlling acceleration We combine the holonomic and nonholonomic servo constraints in (11) and then transform them into a second-order form (12). This transformation helps ensure the mathematical consistency between the system model and the constraints.

[0040] Considering the uncertainty of the wheel-legged robot system to design an adaptive robust control algorithm, when designing the control function τ, the uncertainty existing in the controlled underactuated mechanical system (UMSs) must be fully considered. To this end, we decompose the matrix M, C and matrix G to more accurately handle and compensate for these uncertainties:

[0041]

[0042] In this framework, and are defined as the nominal (or deterministic) components of matrices M, C, and G, while ΔM, ΔC, and ΔG represent the corresponding uncertainty components. Furthermore, all these functions are set to be continuous, thus ensuring the smoothness and reliability of the control process.

[0043] Next, we designed an adaptive robust control strategy of the following form:

[0044]

[0045] and,

[0046]

[0047]

[0048] in,

[0049]

[0050] Here, k1, k2>0, k1, k2∈R, and n=1, 2, 3, 4... are scalar constants.

[0051] parameter Governed by the following adaptive laws:

[0052]

[0053] in ζ i is the i-th component of vector ζ (i=1,2,...,k), L1, L2, L3>0 and L1, L2, L3∈R.

[0054] Based on cooperative game theory, the performance indicators in the control process are constructed, and then the optimal parameters that can balance the two performance indicators are found through the "Pareto optimal" method. These are the optimal values ​​of two important parameters required to control a complex system. Cooperative game theory focuses on exploring the collaborative interaction and effectiveness of multiple participants. Under this theoretical framework, participants form alliances to achieve common goals and negotiate the distribution of the resulting benefits or costs incurred during the cooperation process. We apply cooperative game theory to the control parameter optimization problem of fuzzy underactuated mechanical systems (UMSs), abstract the control parameters as "players" in the game, and construct two performance indicators based on fuzzy set theory. These indicators are converted into cost functions faced by players in the game. Through a carefully designed cooperative game, the Pareto optimal solution is found and the selection of control parameters is optimized. This method can ensure an effective trade-off between multiple performance objectives and promote the optimization of the overall control performance of the system.

[0055] First, a dual-parameter performance index based on uncertain fuzzy information is designed: the performance indexes of k1 and k2 are proposed. s Establish:

[0056]

[0057] Among them, formula (25) can be obtained by D-operation

[0058]

[0059] Here, it can be seen that the performance index J1 reflects the system at the initial time t s It takes into account the deviations in the system state, such as position, speed, etc., and the effects of the control parameters k1 and k2 on these deviations.

[0060] Similarly, by D-operation, we can obtain

[0061]

[0062] It can be concluded that the performance index J2 reflects the error level of the system at infinity, that is, the final error state of the system. It is usually related to the steady-state performance of the system, measuring whether the system can reach or approach the desired state after a long period of operation.

[0063] Next, the optimal design is obtained through the following cooperative game:

[0064] First, find So for any Both

[0065]

[0066] To find We use Lemma 1: Decision is Pareto optimal if there exists ζ∈R with ζ i >0, i=1, 2...M, To satisfy:

[0067]

[0068] Decision-making is Pareto optimal if there exist weighting factors γ1>0 and γ2>0, and γ1+γ2=1, such that

[0069]

[0070] in,

[0071]

[0072] Therefore, we can solve the following constrained optimization problem:

[0073] For any t s and the given D 1,2 ,

[0074]

[0075] To obtain this solution, we apply the partial differential operator to J(k1, k2)

[0076]

[0077] Next,

[0078]

[0079] Among them, k1∈D1, k2∈D2.

[0080] Subsequently, equation (31) can be minimized by solving equations (35) and (36), and the solution is a Pareto optimal solution.

[0081] Based on the above analysis, the minimum performance index is obtained:

[0082]

[0083] Finally, the balance parameter Applied to adaptive robust control to complete uncertainty control of under-actuated mechanical systems.

[0084] The control effect is analyzed and conclusions are drawn at last.

[0085] Compared with the prior art, the present invention has the following beneficial effects:

[0086] 1. Improve control accuracy and stability: By introducing fuzzy set theory to construct a dynamic model with uncertain parameters, this method can effectively deal with the uncertainty of the system's internal parameters that change over time and are limited, and significantly improve the robot's trajectory tracking accuracy in complex terrain and dynamic environments. Combined with the adaptive robust control algorithm, even when encountering external interference or sudden changes in system parameters (such as changes in ground friction and load fluctuations), it can ensure that the robot maintains a stable motion state and avoids large deviations in trajectory tracking, thereby ensuring the efficiency and accuracy of task execution.

[0087] 2. Enhanced environmental adaptability and robustness: The bipedal wheel-legged robot combines the advantages of wheeled and legged robots and performs well in a variety of terrains. The control method proposed in this invention further enhances its environmental adaptability, enabling the robot to respond flexibly in rugged or unknown environments and maintain efficient movement. At the same time, by introducing cooperative game theory to optimize control parameters, the system can automatically adjust the control strategy when facing different working conditions, significantly improving the overall robustness and reducing the risk of system instability.

[0088] 3. Optimize control parameters to maximize performance: Based on the performance indicators constructed by cooperative game theory, the optimal control parameters that can balance the two key performance indicators are found through the "Pareto optimal" method. This process not only ensures that the robot achieves effective trade-offs between different performance goals, but also promotes the optimization of the overall control performance of the system. Compared with traditional methods, the present invention can provide more refined control parameter settings, so that the robot can perform at its best under various operating conditions.

[0089] 4. Improve the tolerance of initial condition deviation: The adaptive robust control algorithm designed by the present invention adopts an innovative closed-form solution and a leaky adaptive law to ensure that the system can maintain high control accuracy and stability when facing initial condition deviation. This feature makes the robot more reliable in practical applications and reduces the risk of control failure caused by uncertainty in initial conditions.

[0090] 5. Promote technological progress and application expansion: The invention not only brings new perspectives and solutions to the control field of bipedal wheeled leg robots, but also provides strong technical support for their application in complex tasks such as field exploration, first aid, and security. As the technology continues to mature and improve, bipedal wheeled leg robots are expected to play an important role in more fields and promote technological progress and development in related industries. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] Figure 1The design flow chart of the example of the present invention;

[0092] Figure 2 The overall structure diagram of the controller of the bipedal wheeled leg robot system of the example of the present invention;

[0093] Figure 3 Schematic diagram of stability simulation of a bipedal wheel-legged robot system according to an example of the present invention. DETAILED DESCRIPTION

[0094] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0095] See also Figure 1-3 The present invention provides a technical solution: a design method of an adaptive robust control wheel-legged robot based on cooperative game theory, which is as follows: comprising the following steps:

[0096] The form of establishing the dynamic model of a bipedal wheeled leg robot system with uncertain parameters specifically includes: considering a dynamic model of a bipedal wheeled leg robot system with uncertain parameters, a control dynamic model of a fuzzy system is described as follows:

[0097]

[0098] Here, t∈R n is time, χ∈R n are generalized coordinates, It's speed. is the acceleration, χ, and Respectively represent the position, velocity and acceleration of the joints in the bipedal wheeled leg robot system. p , σ represents the uncertainty parameter in the mechanical system, which may change over time. The symbol "Σ" is used to define the accuracy range of "σ". Although there is uncertainty, its range of variation is limited. "τ" is used to represent the control input to the system, B∈R n×m is the coefficient matrix of the control input. M, C and G are all continuous, and M(χ,σ,t), including parameters related to inertia, represents the inertia matrix of the system. and G(χ,σ,t) represent Coriolis / centrifugal force and gravity respectively, and F(χ,σ,t) is friction force. Friction effect, as a common nonlinear natural phenomenon, is widely present in various mechanical systems. In order to reduce the adverse effects of friction on the control performance of mechanical systems, many researchers have proposed a variety of friction modeling methods to deeply explore and grasp the essence of this phenomenon. These models include Coulomb friction model, viscous friction model, Stribeck friction model and Dalbe dynamic friction model, each of which is committed to revealing the core mechanism of friction control.

[0099] The dynamic model of the jumping process of the bipedal wheeled robot can be described as follows:

[0100] The kinetic energy of the body is:

[0101]

[0102] The potential energy of the body is:

[0103] V=Mgy (3)

[0104] Lagrangian operator:

[0105]

[0106] Select the joint motor angle α as the generalized coordinate, the generalized force as the torque τ of the four joint motors, and list the Lagrange equation:

[0107]

[0108] Expanding and simplifying formula (5) yields:

[0109]

[0110] The expressions of y1 and y2 are as follows:

[0111]

[0112] The prescribed trajectory in the control process is expressed as a servo constraint imposed on the fuzzy UMS. First, all uncertainties in equation (1) are bounded, and their boundaries are described by known fuzzy sets, which can be expressed as:

[0113]

[0114] where σ i is the i-th term of σ, is i Fuzzy set, ∑i is σ i The domain of μ is compact. i :σ i →[0,1] is The membership function. The tracking control force τ is provided by the actuator, which makes the fuzzy UMS (1) follow the specified tracking trajectory. At this time, the tracking control problem is considered in the framework of constrained motion, and the specified tracking trajectory is regarded as the servo constraint imposed on the fuzzy UMSs.

[0115] Assume that the fuzzy UMS is subject to the holonomic servo constraint in the form of:

[0116] q i (χ,t) = 0, i = 1,2,...,h (9)

[0117] and the non - holonomic servo constraint of m - h in the form of:

[0118]

[0119] where m < n, and we assume that these constraint equations are differentiable. Combine the first - order derivative of (9) with respect to t and (10), and represent the final set of equations in matrix form as:

[0120]

[0121] where A = [A li m×n , c = [c1, c2,..., c m . Then, combine the second - order derivative of (9) with respect to t and the first - order derivative of (10) with respect to t, and represent the result in matrix form as:

[0122]

[0123] where b = [b1, b2,..., b m . Here, equation (11) controls the velocity (12) controls the acceleration We combine the holonomic and non - holonomic servo constraints in (11) and then convert them to the second - order form (12), which helps to ensure the mathematical consistency between the system model and the constraints.

[0124] Considering the uncertainties of the biped wheel - leg robot system, design the adaptive robust control algorithm. When designing the control function τ, it is necessary to fully consider the uncertainties existing in the under - actuated mechanical systems (UMSs) to be controlled. For this purpose, we decompose the matrices M, C, and matrix G to more precisely handle and compensate for these uncertainties:

[0125]

[0126] In this framework, and ​are defined as the nominal (or deterministic) components of matrices M, C, and G, while ΔM, ΔC, and ΔG represent the corresponding uncertainty components. Furthermore, all these functions are set to be continuous, thus ensuring the smoothness and reliability of the control process.

[0127] Next, we designed an adaptive robust control strategy of the following form:

[0128]

[0129] and,

[0130]

[0131] in,

[0132]

[0133] Here, k1, k2>0, k1, k2∈R, and n=1, 2, 3, 4... are scalar constants.

[0134] parameter Governed by the following adaptive laws:

[0135]

[0136] in ζ i is the i-th component of vector ζ (i=1,2,...,k), L1, L2, L3>0 and L1, L2, L3∈R.

[0137] Based on cooperative game theory, the performance indicators in the control process are constructed, and then the optimal parameters that can balance the two performance indicators are found through the "Pareto optimal" method. These are the optimal values ​​of two important parameters required to control a complex system. Cooperative game theory focuses on exploring the collaborative interaction and effectiveness of multiple participants. Under this theoretical framework, participants form alliances to achieve common goals and negotiate the distribution of the resulting benefits or costs incurred during the cooperation process. We apply cooperative game theory to the control parameter optimization problem of fuzzy underactuated mechanical systems (UMSs), abstract the control parameters as "players" in the game, and construct two performance indicators based on fuzzy set theory. These indicators are converted into cost functions faced by players in the game. Through a carefully designed cooperative game, the Pareto optimal solution is found and the selection of control parameters is optimized. This method can ensure an effective trade-off between multiple performance objectives and promote the optimization of the overall control performance of the system.

[0138] First, a dual-parameter performance index based on uncertain fuzzy information is designed: the performance indexes of k1 and k2 are proposed. sEstablish:

[0139]

[0140] Among them, formula (25) can be obtained by D-operation

[0141]

[0142] Here, it can be seen that the performance index J1 reflects the system at the initial time t s It takes into account the deviations in the system state, such as position, speed, etc., and the effects of the control parameters k1 and k2 on these deviations.

[0143] Similarly, by D-operation, we can obtain

[0144]

[0145] It can be concluded that the performance index J2 reflects the error level of the system at infinity, that is, the final error state of the system. It is usually related to the steady-state performance of the system, measuring whether the system can reach or approach the desired state after a long period of operation.

[0146] Next, the optimal design is obtained through the following cooperative game:

[0147] First, find So for any Both

[0148]

[0149] To find We use Lemma 1: Decision is Pareto optimal if there exists ζ∈R with ζ i >0, i=1, 2...M, To satisfy:

[0150]

[0151] Decision-making is Pareto optimal if there exist weighting factors γ1>0 and γ2>0, and γ1+γ2=1, such that

[0152]

[0153] in,

[0154]

[0155] Therefore, we can solve the following constrained optimization problem:

[0156] For any t s and the given D 1,2 ,

[0157]

[0158] To obtain this solution, we apply the partial differential operator to J(k1, k2)

[0159]

[0160]

[0161] Next,

[0162]

[0163] Among them, k1∈D1, k2∈D2.

[0164] Subsequently, equation (31) can be minimized by solving equations (35) and (36), and the solution is a Pareto optimal solution.

[0165] Based on the above analysis, the minimum performance index is obtained:

[0166]

[0167] Finally, the balance parameter Applied to adaptive robust control to complete uncertainty control of under-actuated mechanical systems.

[0168] Figure 2 This is an overall structural diagram of the bipedal wheeled leg robot system controller provided by an example of the present invention.

[0169] Figure 3 The schematic diagram of the stability simulation of the bipedal wheeled leg robot system provided by the example of the present invention shows that the error is reduced. Due to uncertainty and constraints, the tracking error is large. Through the proposed control, the tracking error becomes smaller, which verifies the superiority of the controller designed by the present invention and can provide higher steady-state tracking accuracy and displacement.

Claims

1. A design method for an adaptive robust control wheel-legged robot based on cooperative game theory, characterized by: It includes the following steps: S1. Establish a dynamic model of a fuzzy underactuated system with uncertain parameters, and then write out the dynamic model of the jumping process of the bipedal wheel-leg robot; S2. In the control process, express the specified trajectory as a servo constraint imposed on the fuzzy UMS; S3. Design an adaptive robust control algorithm considering the uncertainty of the wheel-leg robot system; S4. Construct a performance index in the control process based on cooperative game theory, and then find the optimal parameters that can balance the two performance indexes through the "Pareto optimal" method, which are the best values of two important parameters required when controlling a complex system, specifically including the following: First, a dual-parameter performance index based on uncertain fuzzy information is designed: the performance indexes of k1 and k2 are proposed. s Establish: Among them, formula (25) is obtained through D-operation: It can be seen that the performance index J1 reflects the system at the initial time t s The cumulative error to infinity takes into account the deviations of the system state, such as position, speed, etc., and the influence of the control parameters k1 and k2 on these deviations; Similarly, formula (26) is obtained through D-operation Thus, the performance index J2 is obtained, which reflects the error level of the system at infinity, that is, the final error state of the system. It is usually related to the steady-state performance of the system and measures whether the system can reach or approach the desired state after long-term operation; Next, the optimal design is obtained through the following steps of cooperative game: First, find So for any Both have: To find Using Lemma 1: Decision is Pareto optimal if it exists To satisfy: Decision-making is Pareto optimal if there exist weighting factors γ1>0 and γ2>0, and γ1+γ2=1, such that: Among them, Therefore, solve the following constrained optimization problem: For any t s and the given D 1,2 , To obtain this solution, the partial differential operator is applied to J(k1, k2) Next, make Among them, k1 ∈ D1, k2 ∈ D2; Subsequently, equation (31) is minimized by the solutions of equations (35) and (36), and this solution is the Pareto optimal solution; Based on the above analysis, the obtained minimum performance index is obtained: Finally, the balance parameter Applied to adaptive robust control to complete uncertainty control of underactuated mechanical systems; S5. Conduct numerical simulation, observe the results, analyze the control effect, and finally draw a conclusion.

2. The design method of the adaptive robust control wheel-legged robot based on cooperative game theory according to claim 1 is characterized by: The specific form of establishing the dynamic model of the fuzzy underactuated system with uncertain parameters in step S1 includes: Consider a dynamic model of a bipedal wheel-leg robot system with uncertain parameters. The control dynamic model of a class of fuzzy systems is described as follows: Where t∈R n is time, χ∈R n are generalized coordinates, It's speed. is the acceleration, χ, and denote the position, velocity and acceleration of the joints in the bipedal wheeled leg robot system, σ∈Σ∈R p , σ represents the uncertainty parameter in the mechanical system, which changes over time, the symbol "Σ" is used to define the accuracy range of "σ", although there is uncertainty, its range of variation is limited, and "τ" is used to represent the control input to the system, B∈R n×m The coefficient matrix for the control input, M, C and G are all continuous, M(χ,σ,t), including the parameters related to inertia, represents the inertia matrix of the system, and G(χ,σ,t) represent the Coriolis / centrifugal force and gravity, respectively, and F(χ,σ,t) is the friction force; The dynamic model of the jumping process of the bipedal wheel-leg robot is described as follows: The kinetic energy of the body is: The potential energy of the body is: V = Mgy (3) Lagrangian operator: Select the joint motor rotation angle α as the generalized coordinate, and the generalized force is the torque τ of 4 joint motors. List the Lagrange equation: Expand and simplify formula (5) to get: Among them, the expressions of y1 and y2 are as follows:

3. The design method of the adaptive robust control wheel-legged robot based on cooperative game theory according to claim 1 is characterized by: In step S2, expressing the specified trajectory as a servo constraint imposed on the fuzzy UMS in the control process includes: First, all uncertainties in equation (1) are bounded, and their boundaries are described by known fuzzy sets, which are represented as: where σ i is the i-th term of σ, is i Fuzzy set, ∑i is σ i The domain of discourse is compact, μ i :σ i →[0, 1] is The tracking control force τ is provided by the actuator, which makes the fuzzy UMS (1) obey the specified tracking trajectory. At this time, the tracking control problem is viewed in the framework of constrained motion, and the specified tracking trajectory is regarded as a servo constraint imposed on the fuzzy UMSs. Assuming that the fuzzy UMS is subject to the complete servo constraint, its form is: q i (χ,t)=0, i=1,2,...,h (9) And the non-holonomic servo constraint of m - h, whose form is: Among them, m < n. Assume that these constraint equations are differentiable. Combine the first derivative of (9) to t with (10), and express the final set of equations in matrix form as: where A=[A li ] m×n , c=[c1,c2,…,c m ], then, the second-order derivative of (9) to t is combined with the first-order derivative of (10) to t, and the result is expressed in matrix form as: Where b=[b1,b2,…,b m ], where equation (11) controls the speed (12) Controlling acceleration The holonomic and nonholonomic servo constraints in (11) are combined and then transformed into the second-order form (12).

4. The design method of the adaptive robust control wheel-legged robot based on cooperative game theory according to claim 1 is characterized in that: The design of the adaptive robust control algorithm considering the uncertainty of the wheel-leg robot system in step S3 is as follows: When designing the control function τ, it is necessary to fully consider the uncertainties existing in the controlled underactuated mechanical systems (UMSs). For this purpose, decompose the matrices M, C, and matrix G to more precisely process and compensate these uncertainties: In this framework, and is defined as the nominal (or deterministic) components of matrices M, C and G, while ΔM, ΔC and ΔG represent the corresponding uncertainty components. Furthermore, all these functions are set to be continuous, thus ensuring the stability and reliability of the control process; Next, design an adaptive robust control strategy in the following form: And, Among them, Wherein, k1, k2>0, k1, k2∈R, and n=1, 2, 3, 4... are scalar constants; parameter Governed by the following adaptive laws: in, ζ i It is the i-th component of vector ζ (i=1, 2, ..., k), L1, L2, L3>0 and L1, L2, L3∈R.

Citation Information

Patent Citations

  • Control method, device and equipment of wheel-legged robot and readable storage medium

    CN113753150A

  • Wheeled mobile robot control method based on generalized performance measurement

    CN117075604A

Cited By

  • Four-foot cooperative robot preset performance tracking control method based on Nash game

    CN120255336A

  • Self-adaptive robust control method for under-actuated scraper conveyor

    CN122194695A

  • An Adaptive Robust Control Method for Underactuated Scraper Conveyors

    CN122194695B

  • Bridge crane integral sliding mode anti-swing control method based on HJI theory and S-shaped trajectory

    CN122488568A