next up previous
Next: Introduction

Stereo vision based mapping and navigation for mobile robots

Cullen Jennings Don Murray
Department of Computer Science,
University of British Columbia, Vancouver
{jennings, donm}


This paper describes a visually guided robot that can plan paths, construct maps and explore an indoor environment. The robot uses a trinocular stereo vision system to produce highly accurate depth images at 2 Hz allowing it to safely travel through the environment at 0.5 m/s. The algorithm integrates stereo vision, occupancy grid mapping, and potential field path planning techniques to form a robust and cohesive robotic system for mapping and navigation. Stereo vision is shown to be a viable alternative to active sensing devices such as sonar and laser range finders.

Vladimir Tucakov
Wed Dec 4 11:45:59 PST 1996