Sonar and laser based HIMM map building for collision avoidance for mobile robots

This paper presents a sonar and laser-based sensing system for collision avoidance for mobile robots that operate in a particular semi-structured environment: the parking lot area. In this system, 26 sonars are positioned around an omni-directional robot and 2-D laser is located in the front. Histogramic In-Motion Mapping algorithm is applied for map building using data from both the sonars and the 2-D laser. Decisions for collision avoidance are made based on this HIMM map together with the velocity vector of the robot in its Body Fixed Coordinate System. Experimental results show that the robot can operate reliably in a STOP-RESUME manner in a real dynamic parking lot environment.