Autonomous stereo visual guidance and control of mobile robots

Abstract

This paper presents the design and implementation of a stereo visual guidance and control system for mobile robots. A stereo vision system is proposed which is capable of reconstructing 3D map based on observed geometrical features without a priori knowledge on the environments. Specifically, detected edge features in binocular images are processed in Hough space for matching correspondence before 3D reconstruction. The generated 3D map can be effectively updated using visual tracking techniques for matched features together with correspondence matching algorithms for newly-observed features. Moreover, based on the reconstructed 3D map, the desired trajectory can be determined. Thus a visual feedback control law can be synthesized for a robot to follow the desired trajectory. Specifically, angular velocity commands are transmitted to a 8051-based driver module to execute low level control of DC motors and video cameras. Due to the fact that visual tracking is employed to assist in maintaining correspondence information and reduce redundant computation, the proposed system can be operated in real time. The proposed system is successfully validated through experiments in indoor environments.