Mobile Robot Path Planning for Unknown Static Obstacle Avoidance by Improved RRT* Algorithm

Abstract

Unknown static obstacles are a challenging task for mobile robot path planning. When using a global path planning algorithm the resulting path can collide with unknown static obstacles. An Improved Rapidly-Exploring Random Tree (Improved RRT*) was previously proposed for global path planning. The novelty of this current paper is that we propose an unknown static obstacle detection method and combine it with the previously proposed Improved RRT* algorithm. When the robot faces an unknown obstacle, the path planning algorithm finds a new path to achieve the same goal. The mobile robot can pick up and place objects from one station to another. A circle detection method is used for guiding the robot to the correct position related to the station. Experimental results show the correctness and stability of the Improved RRT* algorithm when dealing with path planning tasks with unknown static obstacles. Furthermore, a manipulator is installed on the robot for performing pick and place tasks.