Control method for tracked robot, and tracked robot

Abstract

The invention relates to the field of mobile robot motion control, and discloses a control method for a tracked robot, and the tracked robot. According to the invention, the method comprises the steps: taking the tracked robot as a cascading system which consists of a motor drive system and a vehicle body motion system; constructing an adaptive integral sliding-mode switching function of a variable inclination parameter, and proposing adaptive sliding-mode tracking control based on the equivalent control and switching control according to the adaptive integral sliding-mode switching function;feeding back a drive motor time-varying uncertainty parameter, obtained through the online recognition and the speed of the robot, and a target posture error which is solved in a kinematic model to acontroller of a drive system; decomposing the expected speed of each motor according to the kinematic relation, and achieving the stable motion control of the robot.

Description

technical field [0001] The invention relates to motion control of a mobile robot, in particular to a control method for a crawler robot and the crawler robot. Background technique [0002] With the wide application of agricultural tracked robots (Agricultural Tracked Robot, ATR), higher requirements are put forward for the adaptability, control accuracy and motion stability of the robot system. However, the farmland terrain is complex, the environment is changeable, and the strong coupling and model uncertainty of the ATR system itself make the design and use of the ATR control system more difficult. [0003] For this reason, scholars at home and abroad have conducted in-depth research on it. Some scholars have proposed various trajectory tracking controls, mainly using kinematic models, or models combining kinematics and dynamics. Some scholars have proposed linear feedback methods, PID Control method, calculated moment method, backstepping method, sliding mode control, ne...

AI Technical Summary

Problems solved by technology

Among them, the linear feedback method is a commonly used control method, because the ATR model is nonlinear, so the control accuracy is low; the PID control method, because its control parameters are fixed, the control effect on nonlinear and structurally uncertain systems is poor; calculation The torque method relies on the dynamic model of the controlled object, and the dynamic modeling itself is very complicated and difficult to construct, so the method has little theoretical and practical significance; although the neural network method can overcome the uncertainty and unknown disturbance, but the control algorithm is complicated; fuzzy control does not need to establish an accurate mathematical model, and is suitable for the control of nonlinear time-varying and lagging systems, but the choice of fuzzy rules is not systematic and difficult to adjust online; the virtual control amount is used in the backstepping method , while performing iterative derivation, the controller structure is very complex, and the engineering implementation is difficult

[0004] The complexity of the farmland environment causes many uncertainties in the ATR movement process, such as: parameter perturbation and load disturbance, as well as sensor measurement errors, will cause the ATR trajectory to deviate from the reference path

Conventional control methods are difficult to meet the requirements of high-precision trajectory tracking control

Images