Semi-Automated Emergency Landing Site Selection Approach for UAVs

Abstract

The use of unmanned aerial vehicles (UAV) in military and industry today is becoming more widespread. There are a wide range of UAV models that are functional today. The size of these UAVs can be as small as a hawk and can be as big as a passenger jetliner. It is critical for these UAVs to have contingency plans before flight in case of unexpected situations, such as engine-out events which cause total loss of thrust during flight. An important part of contingency planning is to identify emergency landing sites along the flight path of the UAV. This paper discusses the development of an offline semi-automated approach for finding emergency landing sites in the shape of a rectangular runway to be used in preflight contingency planning. The approach introduces a total of five emergency landing measures and a surface type estimation, which are applied to the identified emergency landing site candidates for their safety assessment. The output is a list of emergency landing site candidates together with their surface type estimates that are ranked from the safest to least safe through a generalized safety score for each surface type. The approach can label the ranked landing site candidates according to their reachability in the presence of wind, given the UAV's altitude and coordinates at the time the total loss of thrust happened and the wind forecast for the area.