Model-free unified tracking and regulation visual servoing of nonholonomic mobile robots

Abstract

In this paper, a 2-1/2-D unified tracking and regulation visual servoing strategy is proposed for a nonholonomic mobile robot system equipped with an onboard camera. Firstly, by comparing the feature points of the unmodeled target object from a reference image with those on the current image and the prerecorded desired sequence of images, a novel composite error vector is defined including both image signals and rotational angles. Subsequently, the chained-form error system dynamics is obtained after utilizing a transformation formed by a new error signal set. Then, adaptive Lyapunov-based techniques are employed to construct online estimate to compensate for constant unknown parameters, and the asymptotically convergent performance is achieved by designing a time-varying continuous controller despite the lack of object model and depth information. Simulation results are collected to investigate the feasibility of the proposed approach.