Neural networks control of a nonholonomic mobile robot with deadzone compensation

This paper presents a control structure designed by backstepping method for nonholonomic mobile robots with deadzone compensation scheme. The backstepping controller makes integration of a kinematic controller and a torque controller without the perfect velocity tracking assumption. The deadzone precompensator using two neural networks(NNs),one to estimate the unknown deadzone and the other to provide adaptive compensation for general nonlinear actuator deadzones of unknown width. And a novel neural network structure is presented for approximation of piecewise continuous functions of the sort that appear in deadzone. The neural precompensator is employed to improve the performance of the backstepping controller. Stability analysis and convergence of tracking errors to zero as well as the learning algorithms for weights are guaranteed with basis on Lyapunov method. Simulations results are provided to show the effectiveness of the proposed approach.