Runtime Performance Verification of Assured Contingency Landing Management for Advanced Air Mobility

Advanced air mobility (AAM) will require increasingly autonomous systems to assure safe on-demand transportation despite the potential for in-flight hazards and failures. This paper presents an Assured Contingency Landing Management (ACLM) framework that innovatively integrates preflight and online planning to enhance safety while minimizing the need for real-time computation. A nominal flight plan and associated risk-categorized landing site database generate and store contingency landing plans before the flight. During flight, ACLM monitors aircraft controllability and destination reachability to rapidly identify any need for an urgent or emergency landing. To react, a viable urgent landing plan is retrieved and executed when available, or a new flight plan is constructed and executed otherwise. This paper describes the realization of ACLM in a small hexacopter uncrewed aircraft system (UAS) use case. Experimentally derived performance, battery, and propulsion system degradation models are used in controllability, reachability, and flight planning. Results show the efficacy and statistical performance of the proposed ACLM architecture in progressively challenging degradation and failure scenarios. The integrated ACLM architecture and its application to a realistic hexacopter and urban map case study are contributions of this work.