Mobile robot trajectory tracking PR control system, method and parameter self-adjustment method

By using the PR control system and method, combined with PRR feedforward control and parameter self-tuning, the noise problem of PD control was solved, and the trajectory tracking accuracy and dynamic response speed of the mobile robot were improved.

CN116088514BActive Publication Date: 2026-11-10SHANGHAI SAGE INTELLIGENT TECH CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202310013221.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-01-05
Publication Date
2026-11-10
Estimated Expiration
2043-01-05

AI Technical Summary

Technical Problem

Traditional PD control methods are prone to data noise in mobile robot trajectory tracking, which leads to a decrease in trajectory tracking accuracy.

Method used

A PR control system and method are adopted, including a PR controller, robot motion equations and walking wheel drivers. By calculating the trajectory error in real time and outputting the desired speed command, the control effect is optimized by combining PRR feedforward control and parameter self-tuning methods.

Benefits of technology

It improves the trajectory tracking accuracy of mobile robots, reduces noise interference, and enhances dynamic response speed and overall control effect.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116088514B_ABST
    Figure CN116088514B_ABST
Patent Text Reader

Abstract

This invention relates to a Proportional-Delay (PR) control system, method, and automatic parameter adjustment method for mobile robot trajectory tracking in the field of mobile robot control technology. The PR controller calculates the trajectory error based on the actual trajectory fed back in real time by the walking wheel driver and the preset desired trajectory, and outputs the desired speed command based on the trajectory error. The robot motion equation receives and decomposes the command from the PR controller and then sends it to the walking wheel driver. The walking wheel driver controls the movement of the mobile robot's walking wheels and feeds back the real-time actual trajectory to the PR controller. The PR control (proportional-delay) method replaces the original PD control method, eliminating the differential link in the traditional PD controller, eliminating noise interference caused by differential calculation, and improving the trajectory tracking accuracy of the mobile robot. In addition, this invention also proposes an automatic adjustment method for PR control parameters based on regression calculation to solve the problem of automatic adjustment of PR controller parameters.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of mobile robot control technology, and in particular to a mobile robot trajectory tracking (PR) control system, method, and parameter self-adjustment method. Background Technology

[0002] Mobile robots plan their trajectories based on the scene conditions and follow the planned routes to complete their mobile tasks. The accuracy of trajectory tracking not only affects the operating efficiency of mobile robots in the scene, but also their safety during movement. Therefore, how to improve the accuracy of trajectory tracking, which is one of the main indicators of mobile robot performance, has always been a key issue of concern in the industry.

[0003] To improve trajectory tracking accuracy, it is necessary not only to plan routes reasonably and scientifically based on the scene conditions and to use high-performance sensors, but also to design more optimized control methods.

[0004] Mobile robot pose control often uses the PD (proportional-derivative) control method. In this method, it is necessary to perform differential calculations on the pose angle, movement position, etc. of the mobile robot. However, the differential calculation of signal data such as angle and position often causes additional signal noise or data noise in practice. This noise can have a significant impact on the final control effect and reduce the trajectory tracking accuracy.

[0005] Therefore, there is an urgent need for a new control method for mobile robots to improve the accuracy of trajectory tracking. Summary of the Invention

[0006] To address the problem of reduced trajectory tracking accuracy caused by data noise in traditional PD control, this invention discloses a mobile robot trajectory tracking PR control system (proportional-delay control), method, and parameter self-adjustment method. The technical solution of this invention is implemented as follows:

[0007] A PR control system for trajectory tracking of a mobile robot includes a PR controller, a robot motion equation, and a walking wheel driver;

[0008] The PR controller calculates the trajectory error based on the actual trajectory fed back in real time by the walking wheel drive and the preset desired trajectory, and outputs the desired speed command based on the trajectory error.

[0009] The robot's motion equations receive and decompose instructions from the PR controller and then send them to the walking wheel drive;

[0010] The walking wheel driver controls the movement of the mobile robot's walking wheels and feeds back the real-time actual trajectory to the PR controller;

[0011] in,

[0012] q(t) = [x(t), y(t), θ(t)] T ;

[0013] q r (t)=[x r (t), y r (t), θ r (t)] T ;

[0014] e(t) = [e x (t), e y t), e r (t)] T ;

[0015] ω r (t)=k p e(t)-k r e(th);

[0016] In the above formulas, t represents time, and q represents time. r Let q(t) be the expected trajectory of the mobile robot at time t, q(t) be the actual trajectory of the mobile robot at time t, and e(t) be the trajectory error of the mobile robot at time t. r (t) represents the desired velocity, k p k is the proportionality coefficient. r θ(t) represents the delay coefficient, h represents the delay time set by the PR controller, T represents the time constant, x(t) and y(t) represent the coordinates of the mobile robot at time t, and θ(t) represents the attitude angle of the mobile robot at time t.

[0017] Preferably, it also includes a PRR feedforward controller (proportional-delay-delay feedforward control);

[0018] The PRR feedforward controller multiplies the difference between the desired trajectory commands of the mobile robot at two consecutive moments by a feedforward coefficient to obtain the feedforward control quantity. The feedforward control quantity output by the PRR feedforward controller is superimposed with the output quantity of the PR controller to form the superimposed desired velocity output to the robot's motion equation.

[0019] in,

[0020] u(t)=k f (q r (th)-q r (t-2h);

[0021] In the formula, u(t) is the feedforward control quantity, and k f This is for feedforward gain.

[0022] A PR control method for trajectory tracking of a mobile robot includes the following steps:

[0023] A1. Establish the PR control system for the mobile robot and initialize the parameters;

[0024] A2, calculate the trajectory error between the expected trajectory and the actual trajectory of the mobile robot;

[0025] A3, input the trajectory error to the PR controller;

[0026] A4, the PR controller outputs the desired velocity command to the robot's motion equation based on the trajectory error;

[0027] A5, the robot's motion equations are decomposed into the desired velocity command and sent to the walking wheel drive;

[0028] A6, the walking wheel drive controls the movement of the walking wheels to complete the movement task.

[0029] Preferably, step A4 further includes PRR feedforward control; the feedforward control quantity output by the PRR feedforward controller is superimposed with the output quantity of the PR controller to form the superimposed desired velocity output to the robot motion equation;

[0030] In step A5, the robot's motion equations are decomposed and superimposed to the desired velocity command, which is then sent to the walking wheel driver.

[0031] in,

[0032] ω r * (t)=u(t)+ω r (t);

[0033] In the formula, ω r * (t) represents the superimposed desired speed command.

[0034] An automatic parameter adjustment method for a PR controller used for trajectory tracking of a mobile robot includes the following steps:

[0035] B1, the PR controller outputs the desired velocity command to the robot's motion equation;

[0036] B2, the robot's motion equations are decomposed into the desired velocity command and sent to the walking wheel drive;

[0037] B3, the walking wheel driver controls the movement of the mobile robot's walking wheels and provides feedback on the actual trajectory;

[0038] B4, calculate the error between the expected trajectory and the actual trajectory;

[0039] B5, input the error into the weight iteration calculation module and the regression calculation module;

[0040] B6, the iterative calculation module obtains the optimal weight value through iterative calculation;

[0041] B7, substitute the optimal weight values ​​into the regression equation in the regression calculation module, calculate the output, and track the error;

[0042] B8: Check if the steady-state tracking error and convergence speed meet the requirements; if they do, output the optimal weight value; if not, change the numerical point value and re-enter B6.

[0043] B9. Select the proportional coefficient and delay coefficient of the PR controller according to the principle of minimizing steady-state error.

[0044] This invention addresses the problem of reduced trajectory tracking accuracy caused by noise from differential calculations in traditional PD control methods. By employing a PR controller for mobile robot trajectory tracking, it improves trajectory tracking accuracy. Furthermore, based on this PR control method, feedforward control is implemented, further enhancing trajectory tracking accuracy. In addition, this invention proposes an automatic adjustment method for PR control parameters based on regression calculation, achieving the technical effect of automatic adjustment of PR controller parameters. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only one embodiment of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a schematic diagram of a PR control system embodiment;

[0047] Figure 2 This is a coordinate system diagram of a mobile robot in an embodiment of a PR control system.

[0048] Figure 3 This is a schematic diagram of a PR controller feedforward control method embodiment;

[0049] Figure 4 A schematic diagram illustrating the principle structure of an embodiment of an automatic parameter adjustment method for a PR controller;

[0050] Figure 5 This is a flowchart illustrating the iterative process of weight values ​​in an embodiment of an automatic parameter adjustment method for a PR controller.

[0051] Figure 6 This is a flowchart of regression analysis in an embodiment of an automatic parameter adjustment method for a PR controller. Detailed Implementation

[0052] The technical solutions of the present invention will now be clearly and completely described with reference to the embodiments and accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] Example

[0054] In one specific embodiment, such as Figure 1 As shown, a PR control system for trajectory tracking of a mobile robot includes a PR controller, a robot motion equation, and a mobile robot. The mobile robot is dual-wheel drive, including two walking wheels, two drive wheels, and a passive wheel for supporting the vehicle body.

[0055] In this system, the PR controller calculates the trajectory error based on the actual trajectory fed back in real time by the walking wheel drive and the preset desired trajectory, and outputs the desired speed command based on the trajectory error.

[0056] The robot's motion equations receive and decompose instructions from the PR controller and then send them to the walking wheel drive;

[0057] The walking wheel driver controls the movement of the mobile robot's walking wheels and feeds back the real-time actual trajectory to the PR controller;

[0058] like Figure 2 As shown, let the coordinates of the robot's center point A at time t be x(t) and y(t), and the angle be θ(t).

[0059] in,

[0060] q(t) = [x(t), y(t), θ(t)] T ;

[0061] q r (t)=[x r (t), y r (t), θ r (t)] T ;

[0062] e(t) = [e x (t), e y (t), e r (t)] T ;

[0063] The expression for the PR controller is: ω r (t)=k p e(t)-k r e(th);

[0064] In the above formulas, q r Let q(t) be the expected trajectory of the mobile robot at time t, q(t) be the actual trajectory of the mobile robot at time t, and e(t) be the trajectory error of the mobile robot at time t. r (t) represents the desired velocity, k p k is the proportionality coefficient. r θ(t) represents the delay coefficient, h represents the delay time set by the PR controller, T represents the time constant, x(t) and y(t) represent the coordinates of the mobile robot at time t, and θ(t) represents the attitude angle of the mobile robot at time t.

[0065] The method for trajectory tracking control of a mobile robot according to this embodiment is as follows:

[0066] A PR control method for trajectory tracking of a mobile robot includes the following steps:

[0067] A1. Establish the PR control system for the mobile robot and initialize the parameters;

[0068] A2, calculate the trajectory error based on the expected trajectory and the actual trajectory of the mobile robot;

[0069] A3, input the trajectory error to the PR controller;

[0070] A4, the PR controller outputs the desired velocity command to the robot's motion equation based on the trajectory error;

[0071] A5, the robot's motion equations are decomposed into the desired velocity command and sent to the walking wheel drive;

[0072] A6, the walking wheel drive controls the movement of the walking wheels to complete the movement task.

[0073] In this embodiment, the PR controller expression does not contain the differential element found in traditional PD controllers. Therefore, this embodiment directly eliminates noise interference caused by differential calculations.

[0074] In a preferred embodiment, such as Figure 3 As shown, the PR control system also includes a PRR feedforward controller.

[0075] The PRR feedforward controller calculates the feedforward control quantity by multiplying the difference between the desired trajectory commands of the mobile robot at two consecutive moments by a feedforward coefficient. The feedforward control quantity output by the PRR feedforward controller is superimposed with the output of the PR controller to form the superimposed desired velocity, which is then output to the robot's motion equation.

[0076] in,

[0077] u(t)=k f (q r (th)-q r(t-2h);

[0078] In the formula, u(t) is the feedforward control quantity, and k f This is for feedforward gain.

[0079] In this implementation, the output values ​​of the PR controller and the PRR feedforward controller are superimposed and used together as the input of the robot's motion equation. The PRR feedforward control can quickly correct trajectory control errors and improve the dynamic response speed of the overall PR control system.

[0080] In step A5, the robot's motion equations are decomposed and superimposed to the desired velocity command, which is then sent to the walking wheel driver.

[0081] in,

[0082] ω r * (t)=u(t)+ω r (t);

[0083] In the formula, ω r * (t) represents the superimposed desired speed command.

[0084] This embodiment further reduces trajectory tracking error through PRR feedforward control.

[0085] Example 2

[0086] In a specific embodiment 2, such as Figure 1-6 As shown, an automatic parameter adjustment method for a PR controller used for trajectory tracking of a mobile robot is presented. This embodiment relies on the basic implementation of Embodiment 1. The principle structure of this embodiment is as follows: Figure 4 As shown.

[0087] Figure 4 In this diagram, S is the Laplace operator, r(s) represents the desired trajectory, and e(s) represents the error; Gc(s) is the PR control system model, and Gp(s) is the controlled system model.

[0088] G c (s)=k p +k r e -hs

[0089] This indicates the PR controller.

[0090] The PR control system model is approximated as a time-delay system, where L is the time delay coefficient of the PR control system.

[0091] The output of the PR control system (i.e., the desired speed command output by the PR controller) y(s) can be expressed as:

[0092]

[0093] In the formula, T is the time constant.

[0094] The general form of the regression equation is:

[0095] z n (x) n =z0+z1x 1 +z2x 2 +z3x 3 +…+z n-1 x n-1 +z n x n ;

[0096] In the formula, z n (x) n These are the regression analysis values; z1, z2, z3, ..., z n This is system data.

[0097] In regression analysis, let:

[0098]

[0099]

[0100] The regression equation can then be further written as:

[0101]

[0102] In the formula, ω n The weights are the weights of the exponential function.

[0103] In regression analysis, the regression equation can be derived as follows:

[0104]

[0105] In the above formula, the weight ω n The optimal value is obtained through iterative calculation and then substituted into the above formula for regression calculation. Specifically, iterative convergence is based on the principle of reducing steady-state error.

[0106] ω n =x n +ψ(x n-1 );

[0107] ψ(x) is an iterative formula. The choice of the iterative formula is common knowledge to those skilled in the art and is not limited in this embodiment.

[0108] The iterative calculation program flow is as follows: Figure 5 As shown: The optimal weight value is obtained through iterative calculation by the iterative calculation module.

[0109] Then as Figure 6 As shown, by substituting the weight values ​​into the regression analysis equation, the optimal K is obtained based on the principle of minimizing steady-state error. p K r Controlling the parameters can ensure improved trajectory tracking accuracy of the robot.

Claims

1. A PR control system for trajectory tracking of a mobile robot, characterized in that, This includes the PR controller, PRR feedforward controller, robot motion equations, and walking wheel drive; The PR controller calculates the trajectory error based on the actual trajectory fed back in real time by the walking wheel drive and the preset desired trajectory, and outputs the desired speed command based on the trajectory error. The robot's motion equations receive and decompose instructions from the PR controller and then send them to the walking wheel drive; The walking wheel driver controls the movement of the mobile robot's walking wheels and feeds back the real-time actual trajectory to the PR controller; in, ; ; ; ; In the above formulas, t represents time, and q represents time. r Let q(t) be the expected trajectory of the mobile robot at time t, q(t) be the actual trajectory of the mobile robot at time t, and e(t) be the trajectory error of the mobile robot at time t. r (t) represents the desired velocity, k p k is the proportionality coefficient. r θ(t) represents the delay coefficient, h is the delay time set by the PR controller; T represents the transpose, x(t) and y(t) are the coordinates of the mobile robot at time t, and θ(t) is the attitude angle of the mobile robot at time t. The PRR feedforward controller multiplies the difference between the desired trajectory commands of the mobile robot at two consecutive moments by a feedforward coefficient to obtain the feedforward control quantity. The feedforward control quantity output by the PRR feedforward controller is superimposed with the output quantity of the PR controller to form the superimposed desired velocity output to the robot's motion equation. in, ; In the formula, u(t) is the feedforward control quantity, and kf is the feedforward gain; The PR control system also includes automatic parameter adjustment; The method for automatically adjusting the parameters is as follows: B1, the PR controller outputs trajectory instructions to the robot's motion equations; B2, the robot's motion equation decomposes the trajectory command and sends it to the walking wheel driver; B3, the walking wheel driver controls the movement of the mobile robot's walking wheels and provides feedback on the actual trajectory; B4, calculate the error between the expected trajectory and the actual trajectory; B5, input the error into the weight iteration calculation module and the regression calculation module; B6, the iterative calculation module obtains the optimal weight value through iterative calculation; B7, substitute the optimal weight values ​​into the regression equation in the regression calculation module, calculate the output, and track the error; B8: Check if the steady-state tracking error and convergence speed meet the requirements; if they do, output the optimal weight value; if not, change the numerical point value and re-enter B6. B9. Select the proportional coefficient and delay coefficient of the PR controller according to the principle of minimizing steady-state error; The PR control method of the system includes the following steps: A1. Establish the PR control system for the mobile robot and initialize the parameters; A2, calculate the trajectory error between the expected trajectory and the actual trajectory of the mobile robot; A3, input the trajectory error to the PR controller; A4, the PR controller outputs the desired velocity command to the robot's motion equation based on the trajectory error; It also includes PRR feedforward control; the PRR feedforward controller output feedforward control quantity and the PR controller output quantity are superimposed to form the superimposed desired velocity output to the robot motion equation; A5, the robot's motion equations are decomposed and superimposed to the desired velocity command, which is then sent to the walking wheel driver; in, ; In the formula, ω r * (t) represents the actual desired speed command; A6, the walking wheel drive controls the movement of the walking wheels to complete the movement task.

Citation Information

Patent Citations

  • Servo system controller with self-adapting fuzzy frictional compensation

    CN101510072A

  • Cascading system theory-based track tracing control method of wheeled mobile robot

    CN110597268A

  • Constant speed travel control device for vehicle

    JP1992046827A