An Exact Euclidean Distance Transform for Universal Path Planning

Abstract

Universal path planning is a major issue in order to facilitate the task required for autonomous systems. Distance transform (wavefront) algorithm is a good path planner and can be used to determine a path near to optimum. We present a new algorithm for universal path planning in exact cell decomposition, using a raster scan method for generating an exact euclidean distance transform (EEDT). This algorithm also find, for every cell in the map, the point sequence to the goal (path smoothing), without post processing. These point sequences are computing by visibility verification between cells.