Competing failure modeling and analysis of phased-mission unmanned aerial vehicles

Unmanned aerial vehicles (UAVs) can be outfitted with ground sensors to offer substantial benefits across a range of applications. Nonetheless, ensuring their reliability in delivering services throughout the designated mission time remains crucial. UAV integrated systems are typically phased-mission systems, which undergo multiple, consecutive, non-overlapping operational phases, each with unique system configurations and component behaviors due to changing tasks and environmental conditions. Components within UAV-integrated systems exhibit statistical dependencies across phases. Furthermore, UAV-integrated systems usually employ relayed communications, where some UAVs (dependent components) rely on other UAVs (relays) to communicate with the base station. If the relay fails, dependent components may increase transmission power to establish a direct link with the base station, but only for a limited and uncertain duration until their power is depleted below the necessary threshold. The competition between isolation and propagated failures is also considered. This paper proposes a combinatorial reliability modeling methodology for phased-mission UAVs subject to competition and random isolation time. The methodology applies to arbitrary types of time-to-failure and time-to-isolation distributions for components. A detailed case study demonstrates the approach. The correctness of the proposed method is verified through Monte Carlo simulations.