An Lrf-Based Obstacle Avoidance Method for Various Robot Sizes

The aim of this study is to develop a robust obstacle avoidance method even when the size of the robot varies. A single Laser Range Finder (LRF) was used for obstacle avoidance. In our previous method, we successfully avoided obstacles even in configurations that would lead to collisions. Specifically, this refers to configurations where collisions would occur if the robot's depth were not considered. We confirmed that the robot did not collide with obstacles under six conditions with varying aspect ratios. We could also avoid obstacles in environments with U-shaped or multiple obstacles. In a similar experimental environment, when using the potential method, the robot stopped, and when using DWA, it avoided a direction that resulted in a detour (taking time to move). The contribution of this paper is that it enables obstacle avoidance regardless of the robot's size and allows for a quick return to the path.