Safe decision making for risk mitigation of UAS

Abstract

This work entails the development of a theoretical framework for the fast and safe reaction of Unmanned Aerial Systems to flight anomalies. The proposed framework uses behavior trees to design behaviors that have safety properties and uses vehicle states to determine best risk control response. A case study is presented where a fixed-wing UAV experiences four types of hazardous scenarios which consist of different types of aircraft faults and external obstacles. The UAV safely reacts by avoiding obstacles when possible, or mitigates the faults by either finding a nearby emergency landing location, or choosing a safe ground impact point far from populated areas. These behaviors have the potential to be used across different aerial robotics platforms, to increase safety amid human error, environmental hazards and aircraft failures.