Online Waypoint Path Refinement for Mobile Robots Using Spatial Definition and Classification Based on Collision Probability

Abstract

A novel online path refinement method is suggested for mobile robots, which refines the given path, including several waypoints using sensor data with a low computational load. The method periodically plans a waypoint path from the current robot position to a certain waypoint on the given waypoint path. While the robot moves along the given path, the path is continuously refined to avoid collisions and local minima caused by unmapped obstacles that are not represented on the map but sensed by sensors. Also, if a shortcut is detected while the robot is moving, the shortcut replaces the existing waypoint path. This online waypoint path refinement (OWPR) method is very practical in terms of computation time and memory size since it is performed regardless of the size of the workspace. To validate the actual performances of OWPR method, various simulations and experiments are conducted. As a result of the simulation and experiment, the maximum execution time of the OWPR algorithm is less than 10 ms, and the generated waypoint path shortened the overall travel distance without collision and falling to the local minima.