Tunnel output feedback tracking control method for wheeled mobile robot

Through the nonlinear extended state observer and tunnel guidance law, the problem of unavailable speed and external interference in the trajectory tracking control of wheeled mobile robots is solved, and the stable evolution of trajectory error within the preset boundary is realized, and the output constraint of the robot is improved. operational reliability.

CN118963337APending Publication Date: 2024-11-15HAINAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410922682.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-10
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The existing wheeled mobile robot trajectory tracking control methods have problems in the face of output constraints, unavailability of speed information, etc., especially in transient and steady state, it is difficult to ensure reliable operation, and it fails to effectively deal with unmeasured speed and external. interference.

Method used

Using nonlinear extended state observer and tunnel guidance law, a dynamic quantization control law is constructed to ensure that the tracking error of the wheeled mobile robot trajectory evolves within the preset behavioral boundary, and the unavailable speed and external perturbation are processed by quantizing the input to achieve omnidirectional Wheel mobile robot tracking control.

Benefits of technology

The stable evolution of the trajectory tracking error of the wheeled mobile robot in transient and steady states within the preset boundary is achieved, parameter uncertainty and external interference are effectively handled, and the operation reliability of the robot under output constraints is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118963337A_ABST
    Figure CN118963337A_ABST
Patent Text Reader

Abstract

The invention discloses a method for tunnel output feedback tracking control of a wheeled mobile robot. The method comprises the following steps: establishing kinematics and dynamics models of the wheeled mobile robot; constructing a non-linear extended state observer based on quantitative input; a tracking error is defined, a kinematics tunnel guidance law is constructed based on an error conversion technology, and it is ensured that the trajectory tracking error of the wheeled mobile robot evolves at a preset behavior boundary; based on a kinematics tunnel guidance law and a non-linear expansion state observer, a dynamic quantitative control law of the wheeled mobile robot is constructed, and trajectory tracking control of the omnidirectional wheel mobile robot is achieved. According to the method, by means of the constructed nonlinear expansion state observer and the designed tunnel guidance law and control law, the wheeled mobile robot can accurately recover the unmeasurable speed and external disturbance.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of robot control, and in particular to a method for tunnel output feedback tracking control of a wheeled mobile robot. Background Art

[0002] In the past few decades, wheeled mobile robots have attracted increasing attention due to their wide applications in military operations, search and rescue, transportation, and other fields. Compared with nonholonomic mobile robots (NMRs), wheeled mobile robots have higher flexibility, reliability, and maneuverability because they can move in any direction without reorientation. Trajectory tracking of wheeled mobile robots is not only a basic control problem, but also a challenging task because they are typical nonlinear systems involving multiple coupled variables, modeling errors, and external disturbances.

[0003] Past research has focused on trajectory tracking control methods for mobile robots, including adaptive control techniques for nonholonomic mobile robots, a robust adaptive controller derived from neural networks, and low-complexity adaptive tracking controllers. However, these methods do not consider the kinematic and dynamic constraints of nonholonomic mobile robots. For wheeled mobile robots, some control strategies include bias-compensated model predictive control (MPC) based on kinematic constraints derived from disturbance observers, MPC based on neural dynamics optimization, and anti-disturbance sliding mode control strategies for wheeled mobile robots based on polar coordinates. Although these controllers can make the tracking accuracy of wheeled mobile robots tend to a small neighborhood of the origin, they still have problems in the face of output constraints and unavailable velocity information.

[0004] In practical applications, ensuring that wheeled mobile robots can operate reliably in both transient and steady-state conditions is a key issue in control design. To address this issue, some studies based on Predicted Performance Control (PPC) have emerged, which are applied to mobile robots, multi-mobile robots and wheeled mobile robots. These studies include a time-accurate tracking controller under constrained conditions, a fault-tolerant PPC scheme for autonomous vehicles with actuator failures, a decentralized formation tracking control strategy for non-holonomic mobile robots with communication constraints, and a vision-dependent formation control strategy. The overshoot of the tracking error cannot be accurately constrained, which is a limitation of the behavioral boundaries of these controllers. Inspired by the above, the present invention aims to design a tunnel output feedback tracking control method for Macnum wheeled mobile robots with unavailable speeds and external disturbances. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a method for tunnel output feedback tracking control of a wheeled mobile robot. In the method, a constructed nonlinear extended state observer and a designed tunnel guidance law and control law are used to enable the wheeled mobile robot to accurately recover unmeasurable speed and external disturbances.

[0006] In order to achieve the above object, the technical solution of the present invention is as follows:

[0007] A method for tunnel output feedback tracking control of a wheeled mobile robot comprises the following steps:

[0008] Establish kinematic and dynamic models of wheeled mobile robots;

[0009] Based on quantized input, a nonlinear extended state observer is constructed;

[0010] Define the tracking error, and construct the kinematic tunnel guidance law based on the error conversion technology to ensure that the trajectory tracking error of the wheeled mobile robot evolves within the preset behavior boundary;

[0011] Based on the kinematic tunnel guidance law and the nonlinear extended state observer, the dynamic quantized control law of the wheeled mobile robot is constructed to realize the trajectory tracking control of the omnidirectional wheeled mobile robot.

[0012] Preferably, the kinematics and dynamics model of the wheeled mobile robot is established in the following steps:

[0013] The kinematic model of the wheeled mobile robot is expressed as follows:

[0014]

[0015] According to the Newton-Euler formula, the dynamic model of the wheeled mobile robot can be derived as follows:

[0016]

[0017] Where R is a rotation matrix, η=[x,y,ψ] T represents the position and yaw angle of WMR, τ represents the control input, ν=[u,v,w] T represents the surge speed, sway speed and yaw rate of the WMR. represents the first-order derivative of v, M represents the control input matrix, f=[f u , f v , f w ] T Represents the total disturbance consisting of internal parameter uncertainty and external disturbance.

[0018] Preferably, based on the quantized input, a nonlinear extended state observer is constructed, including the following:

[0019] The nonlinear extended state observer is formulated as follows:

[0020]

[0021] Where ε is a constant gain and the variable and denote the estimated values ​​of η, ν and f respectively, where Represent x, y, ψ, u, v, w, f respectively u 、f v 、f w The estimated value of Respectively The first-order derivative of , Q(τ) represents the quantized input, R T Represents the transpose of a rotation matrix.

[0022] Preferably, a tracking error is defined, and a kinematic tunnel guidance law is constructed based on an error conversion technology to ensure that the trajectory tracking error of the wheeled mobile robot evolves within a preset behavior boundary, specifically including the following steps:

[0023] Define tracking error, the formula is as follows,

[0024] e=η-η d =[e x ,e y ,e ψ ] T

[0025] where η d =[x d ,y d ,ψ d ] T represents the expected trajectory, e x =xx d ,e y =yy d ,e ψ =ψ-ψ d Denote the position error and heading error respectively. In order to enhance the transient and steady-state behaviors of trajectory tracking, the tracking error e is forced to satisfy the following tunnel preset performance constraints:

[0026]

[0027] in e k (t) and is defined by the following TPP function:

[0028]

[0029] In the above constraints, 0≤δ k , e k,0 =0 is the initial error, where ρ k,0 =ρ k (0) k,∞ =lim t→∞ ρ k (t) are the initial and steady-state values, is the convergence rate, sgn(e k,0 ) represents a symbolic function, when e k,0 = 0, sgn(e k,0 )=0; when e k,0 >0, sgn(e k,0 )=1; when e k,0 <0, sgn(e k,0 )=-1.

[0030] The available variable ζ after equivalent transformation is derived through the following mapping function k , where k = x, y, ψ.

[0031]

[0032] where Φ k is a definition The error transfer function of equation ζ k Taking the derivative, we get:

[0033]

[0034] in, as well as For clarity, the equation Rewritten in vector form:

[0035]

[0036] in, and as well as and and as well as e =[ e x , e y , e ψ ] T according to The derivative of the tracking error is derived as Substitute into the equation Available Define velocity error and filter error: and α e =α f -α where α f is the estimated value of the tunnel guidance law α, and v e and α e Substituting into the equation get:

[0037]

[0038] in represents the estimation error. To ensure that the tracking error e is within the TPP constraint Internal evolution, the tunnel guidance law is derived:

[0039]

[0040] Among them, k 1 is the control gain, which is substituted into the equation The kinematic error dynamic equation can be obtained:

[0041]

[0042] In order to obtain α and its derivatives A smooth signal is introduced with a second-order nonlinear tracking differentiator about α:

[0043]

[0044] in yes The filtered values, γ, β 1 and β 2 is a designed positive constant. Since the second-order nonlinear tracking differentiator is stable, the filtering error is bounded, that is, in and is a normal number.

[0045] Preferably, based on the kinematic tunnel guidance law and the nonlinear extended state observer, a dynamic quantization control law is constructed to realize the trajectory tracking control of the omnidirectional wheel mobile robot, including the following:

[0046] Speed ​​error v e Take the derivation and Bring in:

[0047]

[0048] The designed quantitative control law is as follows:

[0049]

[0050] where k 2 represents the control gain, requiring k 2 >0, substitute the equation Q(τ) into The dynamic equation of dynamic error is obtained:

[0051]

[0052] Based on the above technical scheme, the beneficial effects of the present invention are as follows: the present invention provides a method for tunnel output feedback tracking control of a wheeled mobile robot, establishes a kinematic and dynamic model of the wheeled mobile robot, and defines the trajectory tracking error; based on the quantized input, a nonlinear extended state observer is constructed; based on the error conversion technology, a kinematic tunnel guidance law is constructed to ensure that the trajectory tracking error of the wheeled mobile robot evolves within the preset behavior boundary; based on the kinematic tunnel guidance law and the nonlinear extended state observer, a dynamic quantized control law of the wheeled mobile robot is constructed to realize the trajectory tracking control of the omnidirectional wheeled mobile robot. In order to realize the transient and steady-state evolution of the kinematic tracking error within the tunnel boundary. In order to simultaneously deal with the unavailable speed and unknown total disturbance caused by parameter uncertainty and tire slippage, a nonlinear extended state observer (NLESO) is constructed using quantized input. Through the prescribed performance control method, a tunnel guidance law is formulated to stabilize the kinematic tracking error. Using the anti-disturbance technology, a dynamic quantized control law based on NLESO is proposed for the wheeled mobile robot. The convergence analysis of the proposed closed-loop system is established through the input-state stability and cascade stability theory. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 In a method for tunnel output feedback tracking control of a wheeled mobile robot in an embodiment, the wheeled mobile robot tracks a trajectory within an output limit;

[0054] Figure 2 The present invention is a flow chart of a method for tunnel output feedback tracking control of a wheeled mobile robot in an embodiment;

[0055] Figure 3 is a schematic diagram of an experimental trajectory of a method for tunnel output feedback tracking control of a wheeled mobile robot in an embodiment;

[0056] Figure 4 In one embodiment, a method for output feedback tracking control of a wheeled mobile robot in a tunnel is provided. E -Y E Tracking error in the framework;

[0057] Figure 5is a property of a quantized control input of a method for tunnel output feedback tracking control of a wheeled mobile robot in one embodiment;

[0058] Figure 6 In one embodiment, a method for tunnel output feedback tracking control of a wheeled mobile robot is provided for estimating the velocity of the wheeled mobile robot using a nonlinear extended state observer;

[0059] Figure 7 An embodiment of a method for output feedback tracking control of a wheeled mobile robot in a tunnel using a nonlinear extended state observer to estimate disturbances. DETAILED DESCRIPTION

[0060] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0061] like Figure 2 As shown, this embodiment provides a method for tunnel output feedback tracking control of a wheeled mobile robot, the simulation software uses MATLAB2022, the operating environment is Windows, and specifically includes the following steps:

[0062] Step 1: Mathematically model the wheeled mobile robot (WMR). The kinematic model of the wheeled mobile robot is expressed as follows:

[0063]

[0064] According to the Newton-Euler formula, the dynamic model of this WMR can be derived as follows:

[0065]

[0066] Where R is a rotation matrix. η=[x,y,ψ] T Represents the position and yaw angle of the WMR.

[0067] τ represents the control input. v=[u,v,w] T Represents the surge speed, sway speed and yaw rate of the WMR. represents the first-order derivative of ν. M represents the control input matrix. f=[f u , f v , f w ] T Represents the total disturbance consisting of internal parameter uncertainty and external disturbance.

[0068] Step 2: Consider using a nonlinear extended state observer (NLESO) for the external disturbance f of the WMR. This observer only needs to use the position and orientation information of the WMR to approximate the external disturbance. The NLESO is constructed as follows:

[0069]

[0070] Where ε is a constant gain. and denote the estimated values ​​of η, v and f respectively, where Represent x, y, ψ, u, v, w, f respectively u 、f v 、f w The estimated value of Respectively The first-order derivative of , Q(τ) represents the quantized input, R T Represents the transpose of a rotation matrix.

[0071] Step 3: In order to realize the tracking control of WMR, a tunnel guidance law and a quantized control law are designed. The specific steps are as follows:

[0072] First, define the tracking error as

[0073] e=η-η d =[e x ,e y ,e ψ ] T

[0074] where η d =[x d ,y d ,ψ d ] T represents the expected trajectory. x =xx d ,e y =yy d ,e ψ =ψ-ψ d Denote the position error and heading error respectively. In order to strengthen the transient and steady-state behaviors of trajectory tracking, the tracking error e is forced to satisfy the following tunnel preset performance constraints:

[0075]

[0076] where e k (t) and is defined by the following TPP function:

[0077]

[0078] In the above constraints, 0≤ δ k , e k,0 =0 is the initial error. where ρ k,0 =ρ k (0) k,∞ =lim t→∞ ρ k (t) are the initial and steady-state values, respectively. is the convergence rate, sgn(e k,0 ) represents a symbolic function, when e k,0 = 0, sgn(e k,0 )=0; when e k,0 >0, sgn(e k,0 )=1; when e k,0 <0, sgn(e k,0 )=-1.

[0079] The available variable ζ after equivalent transformation is derived through the following mapping function k , where k = x, y, ψ.

[0080]

[0081] where Φ k is a definition The error conversion function of k Taking the derivative, we get:

[0082]

[0083] in, as well as For clarity, the equation Rewritten in vector form:

[0084]

[0085] in, and as well as and and as well as e ={ e x , e y , e ψ ] T according to The derivative of the tracking error is derived as Then put it into the equation Available Define velocity error and filter error: and αe =α f -α where α f is an estimate of α designed later. e and α e Substituting into the equation get:

[0086]

[0087] in Represents the estimation error.

[0088] In order to ensure that the tracking error e is within the TPP constraint Internal evolution, a tunnel guidance law is derived:

[0089]

[0090] Among them, k 1 is the control gain. Substituting it into the equation The kinematic error dynamic equation can be obtained:

[0091]

[0092] In order to obtain α and its derivatives A smooth signal of α is introduced with a second-order nonlinear tracking differentiator (NLTD):

[0093]

[0094] in yes The filtered value of γ,β 1 and β 2 is a designed positive constant. Since the NLTD system is stable, the filtering error is bounded, that is, in and is a normal number.

[0095] Next, the velocity error v e Take the derivation and Bring in:

[0096]

[0097] Therefore, a quantitative control law is designed as follows:

[0098]

[0099] where k 2 represents the control gain, requiring k 2 >0. Substituting the equation Q(τ) into The dynamic equation of dynamic error is obtained:

[0100]

[0101] Experimental verification analysis: The tunnel output feedback tracking control scheme based on NLESO was verified through actual experimental cases. Figure 3 The control technique presented in was experimentally verified using a wheeled mobile robot with a length of 0.44 m, a width of 0.4 m, and a weight of 8.0 kg on an indoor platform. Figure 4 The main components of the verification platform including OMR, optical capture system, ground control station and WIFI router are shown. Specifically, the optical capture system obtains the real-time position and heading angle of the wheeled mobile robot at a refresh rate of 120Hz. The position, heading angle and control signals are transmitted through the WIFI router at a rate of 5KHz. The tracking error of the closed-loop system is proved to be ISS by the cascade stability theorem. The experimental results verify the effectiveness and feasibility of the proposed tunnel output feedback tracking controller. Figure 1 The wheeled mobile robot used in this article, the experimental site, and its tracking trajectory are shown. Figure 3 Describes the wheeled mobile robot in X E -Y E Expected and actual trajectories in . Figure 4 It can be seen that the tracking error e x , e y and e ψ It can converge within the corresponding constraint space without being affected by model uncertainty and unknown disturbances. Figure 5 The actual control signal input to the quantizer is shown. Figure 6 The recovery performance of the nonlinear ESO for unavailable speeds is presented. Figure 7 is the estimated interference plot of NLESO.

[0102] It should be understood that, although the various steps in the above-mentioned flow chart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the above-mentioned flow chart may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0103] The above are only preferred embodiments of the embodiments of the present application and are not intended to limit the embodiments of the present application. For those skilled in the art, the embodiments of the present application may have various modifications and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the embodiments of the present application shall be included in the protection scope of the embodiments of the present application.

Claims

1. A method for output feedback tracking control of a wheeled mobile robot in a tunnel, characterized in that: The steps include: Establish kinematic and dynamic models of wheeled mobile robots; Based on quantized input, a nonlinear extended state observer is constructed; Define the tracking error, and construct the kinematic tunnel guidance law based on the error conversion technology to ensure that the trajectory tracking error of the wheeled mobile robot evolves within the preset behavior boundary; Based on the kinematic tunnel guidance law and the nonlinear extended state observer, the dynamic quantized control law of the wheeled mobile robot is constructed to realize the trajectory tracking control of the omnidirectional wheeled mobile robot.

2. The method for tunnel output feedback tracking control of a wheeled mobile robot according to claim 1, characterized in that: To establish the kinematic and dynamic model of a wheeled mobile robot, the steps are as follows: The kinematic model of the wheeled mobile robot is expressed as follows: According to the Newton-Euler formula, the dynamic model of the wheeled mobile robot can be derived as follows: Where R is a rotation matrix, η=[x,y,ψ] T represents the position and yaw angle of WMR, τ represents the control input, v = [u, v, w] T represents the surge speed, sway speed and yaw rate of the WMR. represents the first-order derivative of ν, M represents the control input matrix, and f = [f u , f v , f w ] T Represents the total disturbance consisting of internal parameter uncertainty and external disturbance.

3. The method for tunnel output feedback tracking control of a wheeled mobile robot according to claim 2, characterized in that: Based on the quantized input, a nonlinear extended state observer is constructed, including the following: The nonlinear extended state observer is formulated as follows: Where ε is a constant gain and the variable and denote the estimated values ​​of η, v and f respectively, where Represent x, y, ψ, u, v, w, f respectively u 、f v 、f w The estimated value of Respectively The first-order derivative of , Q(τ) represents the quantized input, R T Represents the transpose of a rotation matrix.

4. The method for tunnel output feedback tracking control of a wheeled mobile robot according to claim 3, characterized in that: Define the tracking error, build a kinematic tunnel guidance law based on the error conversion technology, and ensure that the trajectory tracking error of the wheeled mobile robot evolves within the preset behavior boundary. The specific steps include the following: Define tracking error, the formula is as follows, and=η-η d =[and x ,And y ,And ψ ] T where η d =[x d ,y d ,ψ d ] T represents the expected trajectory, e x =ex d ,e y =y i -y d ,e ψ =ψ-ψ d Denote the position error and heading error respectively. In order to enhance the transient and steady-state behaviors of trajectory tracking, the tracking error e is forced to satisfy the following tunnel preset performance constraints: in e k (t) and is defined by the following TPP function: In the above constraints, 0≤ δ k , ek ,0 =0 is the initial error, where ρ k,0 =ρ k (0) k,∞ =lim t→∞ ρ k (t) are the initial and steady-state values, is the convergence rate, sgn(e k,0 ) represents a symbolic function, when e k,0 = 0, sgn(e k,0 )=0; when e k,0 >0, sgn(e k,0 )=1; when e k,0 <0, sgm(e k,0 )=-1. The available variable ζ after equivalent transformation is derived through the following mapping function k , where k = x, y, ψ. where Φ k is a definition The error transfer function of equation ζ k Taking the derivative, we get: in, as well as For clarity, the equation Rewritten in vector form: in, and as well as and and as well as e =[ e x , e y , e ψ ] T according to The derivative of the tracking error is derived as Substitute into the equation Available Define velocity error and filter error: and α e =α f -α where α f is the estimated value of the tunnel guidance law α, and v e and α e Substituting into the equation get: in represents the estimation error. To ensure that the tracking error e is within the TPP constraint Internal evolution, the tunnel guidance law is derived: Where k1 is the control gain, substituting it into the equation The kinematic error dynamic equation can be obtained: In order to obtain α and its derivatives A smooth signal is introduced with a second-order nonlinear tracking differentiator about α: in yes The filtering value of γ, β1 and β2 are designed positive constants. Since the second-order nonlinear tracking differentiator is stable, the filtering error is bounded, that is, in and is a normal number.

5. The method for tunnel output feedback tracking control of a wheeled mobile robot according to claim 4, characterized in that: Based on the kinematic tunnel guidance law and nonlinear extended state observer, a dynamic quantization control law is constructed to realize the trajectory tracking control of the omnidirectional wheel mobile robot, including the following: Speed ​​error v e Take the derivation and Bring in: The designed quantitative control law is as follows: Where k2 represents the control gain, k2>0 is required, and equation Q(τ) is substituted into The dynamic equation of the dynamic error is obtained: