An unmanned aircraft see and avoid algorithm development platform using opengl and opencv

This work presents a see-and-avoid (SAA) system for autonomous Unmanned Aircraft Systems (UAS) and a testbed for evaluating SAA capabilities using simulated cockpit video generated using the computer graphics library OpenGL. The computer vision library OpenCV is used to construct algorithms for background filtering, obstacle detection, and collision avoidance. A distance-based and a distance-agnostic algorithm were developed and evaluated on the proposed testbed. Preliminary results with scenarios with two to four planes are promising.