Adaptive learning controller for autonomous mobile robots

Abstract

An adaptive learning controller is proposed for tracking of nonholonomic mobile robots. It consists of an exponential learning control scheme for velocity dynamics and a reference velocity control scheme for the kinematic steering system. In the adaptive learning controller, the velocity dynamics learning control tracks the reference velocity by learning the inverse function of robot dynamics, while the reference velocity controller stabilises the kinematic steering system to the desired reference model even without assuming an ideal velocity servo. It is shown that all the error signals in the learning control system are bounded and the robot motion trajectory converges to the desired trajectory uniformly and asymptotically. The proposed learning controller is applied to a wheeled mobile robot to demonstrate its feasibility and effectiveness.