基于光流的旋翼无人机障碍物检测方法

针对传统光流法检测障碍物准确性不高、适应性差的问题,提出了一种基于金字塔LK光流的旋翼无人机单目视觉自主障碍物检测方法。该方法通过金字塔LK光流法获取光流信息,结合光流大小和拍摄间距与物距的关系设定自适应阈值,比较光流与阈值的大小并对障碍物进行检测与判定。实验证明:该方法可以有效地检测到障碍物,能满足无人机在多种条件下的避障需求,具有较高的准确性、鲁棒性和适应性。

Obstacle Detection Method Based on Optical Flow for Rotor UAV

To solve the problem of traditional optical flow algorithm's poor accuracy and adaptability for Rotor UAV in detecting obstacle,a monocular autonomous obstacle detection method based on Pyramid LK optical flow is proposed.In the method,optical flow is obtained by Pyramid LK optical flow algorithm.The adaptive threshold is determined by optical flow,photographic distance and object distance.Obstacles are detected by comparing optical flow with the threshold.Experimental results show that the proposed method can detect obstacles effectively,while holding a higher accuracy,robustness and adaptability,which can meet the requirements of UAV obstacle avoidance in various conditions.