Stereo-vision-based Collision Avoidance Simulation

Because of safety concerns with unmanned aerial vehicles (UAV's), an effective collision avoidance algorithm is highly desirable on such vehicles. In this research, a stereo-vision distance-measurement-based UAV collision avoidance system for autonomous high-speed UAV's is proposed. We simulate the algorithm using a test bed consists of an OpenGL (Open Graphic Library) simulator of cockpit video and OpenCV (Open Source Computer Vision Library)-based image processing. The testing and results shows that the proposed algorithm works satisfactorily in both controlled and uncontrolled scenarios.

[1]  Quan Pan,et al.  A vision based sense and avoid system for small unmanned helicopter , 2015, 2015 International Conference on Unmanned Aircraft Systems (ICUAS).

[3]  N. Neogi,et al.  Color Optic Flow: A Computer Vision Approach for Object Detection on UAVs , 2006, 2006 ieee/aiaa 25TH Digital Avionics Systems Conference.

[4]  James Liu,et al.  Obstacle Avoidance System for UAVs Using Computer Vision , 2015 .

[5]  J. Mrovlje,et al.  Distance measuring based on stereoscopic pictures , 2008 .

[6]  Mingyue Ding,et al.  A path replanning algorithm based on roadmap-diagram , 2004, Fifth World Congress on Intelligent Control and Automation (IEEE Cat. No.04EX788).

[7]  Rodney A. Walker,et al.  Computer Vision Based Collision Avoidance for UAVs , 2005 .