Hybrid Visual Servo Trajectory Tracking of Wheeled Mobile Robots

Abstract

A hybrid visual trajectory strategy is developed for wheeled mobile robots equipped with onboard vision systems, wherein the 2.5-D visual servoing framework is utilized to enhance trajectory tracking behavior and help to retain visual objects in horizon of the camera. First, according to the current image, the reference image, and the desired images sequence, compound system errors are constructed by both image features and robot orientation. Subsequently, the open-loop error dynamics can be acquired after introducing an error transformation. On this basis, an adaptive controller is developed to achieve the visual servo tracking objective, where the feature depth is compensated online by a parameter updating mechanism. As demonstrated by Lyapunov techniques and Barbalat's Lemma, the proposed visual trajectory tracking controller makes the system errors converge to zero asymptotically in spite of the unknown scene depth. Comparative simulation results are provided to validate the performance of the presented strategy.