Disturbance Observer Based Path Following Control For Wheeled Mobile Robot

As an intelligent system of highly autonomous capability,mobile robot can replace human to perform various of difficult tasks,such as military affairs gumshoe,mine sweeping for removing danger,and materiel transportation of industrial automation.It is becoming more and more widely used in modern society.Most of these applications can simply be attributed to path following control problems.At the same time,when mobile robot performs path following control tasks,external disturbance reduces the accuracy of path following control.In this paper,a mathematical model of a mobile robot is established,and a nonlinear model predictive control method is proposed.The disturbance observer method is used to suppress the influence of the unknown disturbance on the dynamic system.The specific research contents are as follows:The system model is the basic tool for the control system design.The kinematic model and dynamic model of the wheeled mobile robot are the basis for solving the problem of path following control.For this reason,this paper analyzes the nonholonomic constraint equations of the wheel and the ground of the mobile robot,and establishes the mathematical model of the wheeled mobile robot under ideal conditions and non-ideal conditions.Through modeling and analysis,the mobile robot system belongs to a nonlinear underactuated system with non-holonomic constraints.Nonlinear model predictive control(NMPC)strategy is established to deal with nonlinear dynamics and constraints in this paper.The mathematical description and the involved open-loop optimization problem are given.In order to ensure the recursive feasibility and asymptotic convergence of NMPC strategy,the terminal set,terminal control law and terminal penalty are solved based on polytopic description of nonlinear systems and linear matrix inequality.In the actual robot control system,because the velocity and angular velocity can not directly affect the system,the integral sliding mode controller is designed to follow the ideal velocity and angular velocity.The voltage of DC motor is obtained to control the velocity and angle of the wheeled mobile robot.The simulation is performed to verify the rationality of the control structure.In the actual following control of mobile robot,there are often a variety of uncertainties,such as unmodeled dynamics,parameter perturbations,external disturbances and so on.These uncertainties can degrade system performance and even cause system instability.Therefore,this thesis uses disturbance observer based control to estimate and to deal with the uncertainty in the system.First of all,the disturbances can be divided into two types: matched or unmatched disturbances.For the matched disturbances,a nonlinear observer is designed.For the unmatched disturbances,a terminal sliding mode observer is designed.Both of them can estimate and eliminate the effect of disturbances on system dynamics.Through simulation,it is verified that the disturbance observer based NMPC can effectively eliminate or reduce the influence of disturbances and ensure the mobile robot's convergence to the reference path.