Reliability Assessment of Mid-Air Collision Avoidance System for Unmanned Aircraft Systems

Abstract

Unmanned Aircraft System (UAS) applications include Intelligence, Surveillance, Target Acquisition, and Reconnaissance (ISTAR) missions over the battlefield, as well as those military and civilian missions that are too “dull, dirty, or dangerous” for humans piloting manned aircraft. However, to faultlessly integrate UAS vehicles into the U.S. National Airspace System (NAS), the Federal Aviation Administration (FAA) requires “an equivalent level of safety, comparable to see-and-avoid requirements for manned aircraft.” In other words, an “ownship” unmanned aerial vehicle (UAV) must be capable of sensing intruder aircraft that are both cooperative (i.e., aircraft equipped with a transponder) and noncooperative [i.e., air traffic that either cannot or choose to not communicate with Air Traffic Control (ATC) and other air traffic, including, crop sprayers, sail planes, gliders, hot air balloons, and flocks of birds], as well as executing autonomous maneuvers to avoid intruder aircraft in mid-air. The University of North Dakota has been actively developing enabling airborne sense and avoid (SAA) technology under a series of projects funded by the Department of Defense (DoD). A Local Airspace Surveillance Terminal (LAST) based on the Automatic Dependent Surveillance – Broadcast (ADS-B) transceiver was designed for recording the flight activities in the Grand Forks regional airspace. Approximately 12710 flight hours of flight data, including both UND Aerospace and non-UND Aerospace flights, was recorded over a span of six weeks. The recorded data was used to characterize the local airspace so that nominal flight paths could be generated for realistic simulations. To prove the robustness of an airborne SAA prototype system also based on the ADS-B transceiver software-in-the-loop (SWIL) and hardware-in-the-loop (HWIL) simulation environments have been developed, particularly for the reliability assessment of a mid-air collision avoidance algorithm. This paper depicts the ADS-B data gathering, post processing, and aircraft encounter modeling methods along with preliminary simulation results.