Correspondence and clustering methods for image-based wing-tip collision avoidance techniques

Ramp and taxiway regions are the most congested areas of an aerodrome. High levels of activity are associated with these areas and aircraft are frequently surrounded by several types of obstacles, including other aircraft and vehicles. This paper discusses the correspondence and clustering techniques of a stereo vision system that can be installed on board an aircraft to detect and track generic obstacles around the aircraft during ground manoeuvres.