Vision-based Obstacle Avoidance Algorithm for Mobile Robot

Abstract

The obstacle avoidance technology of mobile robots is the key to affect the reachable space and working ability of mobile robots, and is also a research hotspot in the field of robots. In this paper, under the environment of ROS (Robot Operating System) platform environment, aiming at the shortcomings of traditional mobile robot obstacle avoidance methods, an obstacle avoidance method based on machine vision for dynamic obstacle detection is proposed. In the aspect of vision, this method uses the improved object detection algorithm based on YOLO-v3 to detect dynamic obstacles in real time, and combined with the dynamic window approach (DWA) to improve the local path obstacle avoidance algorithm. So that it can avoid sudden mobile obstacles after detection, increase the ability of local dynamic obstacle avoidance, and improve the local obstacle avoidance performance of mobile robots. Finally, the feasibility and effectiveness of the algorithm are verified in the actual environment.