A detection and relative direction estimation method for UAV in sense-and-avoid
Zhaowei Ma, Tianjiang Hu, Lincheng Shen, Weiwei Kong, Boxin Zhao · 2015
This article mainly focuses on the vision-based obstacle avoidance of unmanned aerial vehicle (UAV). Detecting the obstacle and controlling the UAV to avoid it are two key modules in the sense-and-avoid system. We adopt a bio-inspired human vision algorithm called saliency method to realize the automatic detection. Furthermore, this article proposes a reference yaw angle calculator algorithm based on the geometrical relationship between the UAV and camera. The UAV aims to perform the yaw maneuver to avoid the obstacle in this article. The yaw angle regulating process is achieved via a PD controller operating as the low level control scheme. Finally, proposed methods are verified using the AR. Drone and ROS-based scheme. The experimental results show that the proposed method is a powerful candidate for vision based avoidance in the aerial environment.