Safe Gap based (SG) reactive navigation for mobile robots

Abstract

In this paper a new real-time collision avoidance approach for mobile robots moving in dense and cluttered environments is presented. Our method is to adapt the Closest Gap (CG) Navigation technique, which we have proposed in an earlier work [1], to generate safer and smoother robot trajectories. This is achieved by introducing an additional step in analyzing the structure of obstacles surrounding the robot, and generating instantaneous subgoals which drive the robot towards a collision-free area. The performance of this approach as compared with the CG method is demonstrated with simulation and experimental results.