Neural Dynamics Based Motion Control For Nonholonomic Mobile Robot

This paper considers nonholonomic autonomous wheeled mobile robot as the control object and in view of both theory and application, studies the control problems focusing on trajectory tracking and point stabilization, which are one of the most important issues on nonholonomic mobile robot motion control. The final objective of this paper consists in constructing a nonholonomic mobile robot motion control system composed of trajectory tracking and point stabilization controllers, which are stable and effective in theory and are practical and feasible in application.Firstly, as for trajectory tracking control, this paper establishes posture error model denoted by Cartesian coordinates in local coordinates and proposes a novel tracking control approach for nonholonomic mobile robot by integrating the neural dynamics model into a backstepping technique, and an adaptive neuron model is presented to make robust tracking. It resolves the speed jump problem existed in some previous tracking controllers. The stability of the control system is guaranteed by a Lyapunov stability theory.Then, as for point stabilization control, this paper establishes posture error model denoted by Polar coordinates in global coordinates and makes use of the inherent discontinuousness of the closed-loop system model to proposes a novel tracking control approach for nonholonomic mobile robot by integrating the neural dynamics model into a discontinuous time-invariant technique. Through analysing the local uniform asymptotical stability at origin under the proposed point stabilization control law, the scale of control parameters is confirmed. By Lyapunov candidate function method, this paper concludes that the closed-loop system is globally uniformly asymptotically stable at origin.Finally, utilizing SmarR0B2 nonholonomic middle-size autonomous wheeled mobile robot and by means of software, this paper compiles program to implement nonholonomic mobile robot motion control system. Simulation and implementation results validate that the trajectory tracking and the point stabilization control laws proposed in this paper are effective.