Stereo vision based obstacle avoidance strategy for quadcopter UAV

Abstract

This paper presents a visual control method of a small UAV in which binocular stereo vision and ultrasonic system are fused to implement obstacle avoidance strategy. This combination of sensors allows us to get a complete characterization of the environment and describe obstacle avoidance. It means that stereo vision system can detect obstacles around UAV and calculate a new path by RRT-algorithm. The ultrasonic sensor is introduced to compensate for the situations where something may lead to the failure of the obstacle detection. The method presented in this paper can improve the automatic level of the flying control system. Experiments on obstacle avoidance demonstrate the effectiveness of the proposed method and the performance of the UAV platform.