Visual tracking of mobile robots with both velocity and acceleration saturation constraints

Abstract

In this paper, an adaptive visual tracking controller is proposed for the nonholonomic mobile robot with an onboard monocular camera, wherein both the velocity and acceleration saturation constraints can be guaranteed by properly selecting the control parameters during the visual servoing process. Specifically, the virtual velocity-level controller is firstly designed based on the first-order filter, by which the bounds of the virtual velocity signals can be estimated. Then, according to the finite-time control theory, the actual acceleration-level controller is developed to track the virtual velocity input in a finite time. For the unknown distance between the plane of feature points and the origin of the global frame due to the lack of the image depth, an adaptive updating law is designed and also used for the parameter identification. The global asymptotical stability of the closed-loop system is proven using Lyapunov techniques with Barbalat’s lemma. The calculations for the upper bounds of velocity and acceleration signals are given. In addition, some tips for the control gains tuning in practical experiments are summarized. Experimental results are provided to validate the effectiveness of the proposed controller.