Decentralised fault detection and diagnosis in navigation systems for unmanned aerial vehicles

Autonomous unmanned aerial vehicles (UAVs) are a technological phenomenon sweeping the world stage. Full autonomy implies that the guidance and navigation system employed must exhibit the highest level of integrity. This paper looks at the parity space fault detection and diagnosis (FDD) methods, and its applicability in fully autonomous guidance and navigation systems in a decentralised system architecture. Using the existing work as a starting point this paper identifies the effectiveness of these methods when applied to situations where both the hardware and analytical redundancy exist. One of the most important issues in FDD in navigation systems using redundant sensors relates to the integrity of the solution processing architecture. This has motivated the development of multiple FDD solutions running on numerous separate processors in a decentralised computing network. Typical solutions to this problem are based on decentralised or multiple Kalman filters running in parallel. This paper addresses the use and merits of the information filter form of the Kalman filter in a fully decentralised FDD framework