Adaptive tracking in mobile robots with input-output linearization

Abstract

This paper proposes a neural network adaptive controller augmentation to a computed torque proportional plus derivative (PD) controller, to guide a nonholonomic mobile robot during trajectory tracking. Path following is done defining a point to follow (look-ahead control) and using input-output linearization. A dynamic inversion neural network controller is responsible for tracking error reduction and adaptation to unmodeled external perturbations. The adaptive controller is implemented through a hidden layer feed-forward neural network and has its weights realtime updated. The stability of the whole system is analyzed using Lyapunov theory, and the control errors are proved to be ultimately bounded. Simulation results are also presented, which demonstrate the good performance of the proposed controller for trajectory tracking under external perturbations.This paper proposes a neural network adaptive controller augmentation to a computed torque proportional plus derivative (PD) controller, to guide a nonholonomic mobile robot during trajectory tracking. Path following is done defining a point to follow (look-ahead control) and using input-output linearization. A dynamic inversion neural network controller is responsible for tracking error reduction and adaptation to unmodeled external perturbations. The adaptive controller is implemented through a hidden layer feed-forward neural network and has its weights realtime updated. The stability of the whole system is analyzed using Lyapunov theory, and the control errors are proved to be ultimately bounded. Simulation results are also presented, which demonstrate the good performance of the proposed controller for trajectory tracking under external perturbations.