Occupancy-Elevation Grid Mapping from Stereo Vision

This paper proposes an alternative environment mapping method for accurate robotic navigation based on 3D information. Typical techniques for 3D mapping using occupancy grid require intensive computational workloads to both build and store the map. We introduce an Occupancy-Elevation Grid (OEG) mapping based on visual range data, which is a discrete mapping approach where each cell represents the occupancy probability, the elevation of the terrain and its variance. This representation allows a mobile robot to know if a space in its environment is occupied by an obstacle and the elevation of such obstacle, thus, it can decide if it is possible to traverse the obstacle. The resulting maps allow the execution of tasks like decision making for autonomous navigation, exploration, localization and path planning. Experiments carried out with a real mobile robot equipped by a stereo vision system demonstrate that the proposed approach yields useful maps for autonomous robust navigation.