Abstract
Wheeled mobile robots are known as one of nonholonomic constrained systems and many control strategies have been proposed. As one topic of them, image-based control of mobile robots has been paid much attention because of its robustness against modeling or calibration errors. In this paper, two image-based path following control methods are proposed to follow a target curve marked on the ground. One is a basic controller based on a Lyapunov's direct method and the other is designed by an inverse optimal approach and has robustness to a class of uncertainties. Some simulations and experiments are performed to show their performance.
