Planar region alignment based visual regulation for mobile robot

It has been proved that, based on planar region alignment, camera ego-motion motion can be robustly estimated with a multistage approach in [24]. Based on this observation, a visual regulation approach based on planar region alignment is proposed. This paper is composed of two parts. In the first part, the image motion model of planar point is reviewed first, which will be used for camera motion estimation. Then a simple least square method for computing homography matrix H is introduced. The basic procedure of multistage motion estimation based on planar region alignment is summarized. In the second part, the kinematic model of monocular camera based mobile robot system is deduced with a constrained camera-robot configuration, the visual regulation control law is derived and the stability of the close loop system is analyzed in the sense of Lyapunov stability theory. The motion estimation experiment results are presented with real images, and simulation results show the convergence of the proposed visual regulation.