Robust Control of Mobile Robots on Rough Terrain

Abstract

A wheeled mobile robot is one of the nonholonomic constrained systems and has been studied intensively. An unmanned ground vehicle is a kind of wheeled mobile robots and is strongly needed in Japan because of decrease and aging of agricultural workers. In this paper, inverse optimal control with path regeneration is applied to the path following problem of a nonholonomic mobile robot and an unmanned ground vehicle. The designed controller has a stability margin which guarantees robustness to uncertainties on the input. The robustness has a possibility of superior performance on rough terrain where disturbances to the wheels affect the control accuracy. The efficiency of the proposed control system is confirmed by some simulations and experiments, in which a camera and a three-axial angle sensor are used to obtain the position and attitude.