Path Planning for Autonomous Mobile Robot Based on Safe Space

Abstract

Path planning is essential for the autonomous navigation of mobile robots. Researchers have been working on ensuring the safe navigation of mobile robots; however, it is impossible to secure the absolute safety of a mobile robot without environmental information. Nevertheless, passive safety of a mobile robot can be secured. With the aim of ensuring safe path planning of a mobile robot, a safety space is proposed in this work by using the parameters of stopping distance and hazard point. Mobile robots should formulate path plans to bypass crossroads or corner areas where their field of view is limited, and they should also be capable of reducing their movement speed to secure safe driving. We demonstrate through extensive simulations that the developed SGPP (safe global path planning) method outperforms the classical A* and PRM (probabilistic roadmap) algorithms. It improves the navigation time and length of the robot path in comparison to the PRM algorithm and reduces the navigation time by up to 26% compared to the classical A* algorithm for safe navigation. In addition, results of an experiment conducted on a real robot show that the SGPP method finds a safe path with limited velocity for safe navigation.