Embedded estimation of fault parameters in an unmanned aerial vehicle

In this paper, we present a model-based approach for estimating fault conditions in an aircraft. We formulate fault estimation as a convex optimization problem, where estimates are obtained by solving a constrained quadratic program (QP). A moving horizon framework is used to enable recursive implementation of the constrained QP of fixed size. The estimation scheme takes into account a priori known monotonicity constraints on the faults. Monotonicity implies that the fault conditions can only deteriorate with time. We validate the proposed estimation scheme on a detailed nonlinear simulation model of the Aerosonde unmanned aerial vehicle (UAV) in the presence of winds and turbulence. An excellent performance of the developed approach is demonstrated.