Automated Failure Effect Analysis for PHM of UAV

Abstract

The ASTRAEA project is a £32M UK initiative to develop the safety case for unmanned aerial vehicles flying in commercial airspace. It is addressing both the issue of what needs to be covered by such a safety case, and how such a safety case can be constructed efficiently. One of the key areas within the remit of ASTRAEA is that of generating diagnostics capable of correctly identifying the causes of all possible failures of the vehicle.This paper describes how model-based simulation can be employed to automatically generate the systemlevel effects of all possible failures on systems within the aircraft. The results of the simulation can be used in several ways. They can be used to produce a system-level FMEA for aircraft systems. They can be used to identify the sensors necessary to discriminate remotely between different failures on the aircraft. Once a set of sensors have been chosen for placement on the vehicle, the simulation results can also be used to generate diagnostic and prognostic software for deployment on the vehicle. Using the automated safety analysis software developed on the ASTRAEA project is more efficient than doing the same work without the software, and also provides a guaranteed level of performance.

Automated FMEA based diagnostic symptom generation

“…The generated set of symptoms to be monitored can be augmented by further symptoms that compensate for the limitations of the chosen modelling. In our solution, all of the symptoms are included in an existing proprietary Bayesian network based diagnostic system developed by BAe Systems [19]. The process of using qualitative symptoms within this system is outside the scope of this paper, however several key features are provided in section 6 of this paper.…”