An efficient neural controller for a nonholonomic mobile robot

In this paper, a novel neural network based controller is developed for real-time fine motion control of a nonholonomic mobile robot with completely unknown robot dynamics and under unmodeled disturbance. By taking advantage of the robot regressor dynamics that express the highly nonlinear robot dynamics in a linear form in terms of the robot dynamic parameters, the neural network consists of a single layer feedforward structure, and the learning algorithm is computationally efficient. Unlike previous works that use a typical backstepping velocity planner as the control input, a novel neural dynamics based velocity planner is used as input. The stability of the proposed control system and the convergence of tracking errors to zero are rigorously proved using a Lyapunov theory. The fine control of mobile robot is achieved through the on-line learning of the neural network without any off-line learning procedures. The effectiveness and efficiency of the proposed controller is demonstrated by simulation studies.