Looming aircraft threats: Shape-based passive ranging of aircraft from monocular vision

Abstract

This paper proposes new techniques for aircraft shape estimation, passive ranging, and shapeadaptive hidden Markov model filtering which are suitable for a monocular vision-based noncooperative collision avoidance system. Visionbased passive ranging is an important missing technology that could play a significant role in resolving the sense-and-avoid problem in unmanned aerial vehicles (UAVs); a barrier hindering the wider adoption of UAVs for civilian applications. The feasibility of the proposed shape estimation, passive ranging and shape-adaptive filtering techniques is evaluated on flight test data. Show more