Perception and navigation of mobile robots

Abstract

The authors present a method of navigation for mobile robots, based on perception of the unknown environment, that is similar to a blind man's behavior. The perception uses data from different types of sensors, combined in a probabilistic occupancy map of the close surroundings of the robot. A global map of the workspace is constructed and updated at each step by the navigation algorithm, using the latest perception. The navigation algorithm is suited for indoor as well as outdoor applications. It computes the optimal path based on the last updated map of the workspace and the present position estimate of the robot with respect to the goal position. The algorithm consists of two stages: global path planning and local obstacle avoidance.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>