Research and Hardware Implementation of Binocular Vision Obstacle Avoidance for UAV

Abstract

AbstractWe analyze the advantages and disadvantages of the current mainstream UAV obstacle avoidance solutions and focus on the UAV binocular vision obstacle avoidance method for UAV flight safety. A binocular vision hardware system based on the principle of binocular stereo vision and combined with OpenCV is built to realize the obstacle detection function of UAV binocular vision obstacle avoidance. The BM and SGBM algorithms are chosen as the main strategies for the stereo matching step in the obstacle avoidance method, and the Canny operator edge detection is used to constrain the stereo matching process in the system. The experimental results show that the UAV binocular vision obstacle avoidance method has a smooth and clear output parallax map and good obstacle ranging effect after the Canny operator edge detection constraint in the system, and the detection results apply to the subsequent UAV obstacle avoidance behavior, which can effectively guarantee the UAV flight safety.KeywordsUAVObstacle avoidanceBinocular visionStereo matching