Real-time obstacle detection system for high speed civil transport supersonic aircraft

The High Speed Civil Transport (HSCT) supersonic commercial aircraft under development by National Aeronautics and Space Administration (NASA) and its partners is expected to include an eXternal Visibility System (XVS) to aid the pilot's limited view through their cockpit windows. XVS obtains video images using high resolution digital cameras mounted on the aircraft and directed outside the aircraft. The images captured by the XVS provide an opportunity for automatic computer analysis in real-time to alert pilots of potential hazards in the flight path. The system is useful to help pilots make decisions and avoid air collision. In this paper, we describe the design, implementation, and evaluation of such a computer vision system. Using this system, real-time image data was recently obtained successfully from night tests conducted at NASA Langley Research Center. The system successfully detected and tracked translating objects in real-time during the night test. The system is described in detail so that other researchers can easily replicate the work.