Experimental Verification of Path Planning with SLAM

Abstract

Simultaneous localization and mapping (SLAM) is a very popular technique which is used to develop autonomous mobile robots. We have developed an autonomous mobile robot that can perform the SLAM based solely on information it gathers with a laser range finder. Path planning using the A* algorithm is proposed to help the robot determine the shortest path while avoiding obstacles and minimizing travel distance and rotation. In addition to the standard eight adjacent cells present in conventional A* algorithms, the proposed path planning method allows the eight cells that may be reached via the knight move to be defined as additional adjacent nodes. As a result, the achieved path is smoother than those obtained via more conventional methods, as has been experimentally verified.