A study of improvement of D~* algorithms for mobile robot path planning in partial unknown environments

摘要

Purpose - The purpose of this paper is to improve the D~* algorithm which has been used usually in robotics for mobile robot navigation in unknown or dynamic environments. Design/methodology/approach - First, the model of 2D workspace with some obstacles is expressed in regularity grids. The optimal path is planned by using the improved D~* algorithm by searching in the neighbor grid cells in 16 directions. It makes the robot that the smallest turning angle drops toπ/8. The robot moving angle discrete precision is raised and the unnecessary cost of path planning is reduced so the robot motion path is smoother. Then, the improved D~* algorithm is simulated in MOBOTSIM software environment and is tested by the WiRobotX80 mobile robot. Findings - To search in the neighbor grid cells in 16 directions instead of eight directions by using D~* algorithms for path planning. Research limitations/implications - The map should be expressed in regularity grids. Originality/value - The improved D~* algorithm is effective and it can result in a higher quality path than the conventional D~* algorithm at the same map environment.