Gliding Footprint Calculation Method for Aircraft with Thrust Failure Based on Six-Degree-of-Freedom Flight Envelope and Back-Propagation Artificial Neural Network

Abstract

In hostile environments, engine damage is of particular concern since the engine is the only component to generate thrust that affects survivability. For an aircraft suffering thrust failure, the forced landing sites should be identified within the gliding footprint, which is the reachable region on the ground. This paper proposes two calculation methods to obtain the gliding footprint by finding a series of boundary points with maximum gliding distance around the aircraft. Method 1 models the thrust-failed aircraft with six-degree-of-freedom (6-DOF) flight dynamics and adopts a novel 6-DOF unpowered-flight envelope to characterize its maneuvering capabilities. Given the initial altitude when thrust failure occurs, Method 1 determines all feasible gliding distances around the aircraft based on the constructed 6-DOF flight envelope and selects the landing points of maximum gliding distances along different radial directions as the boundary points. Method 2 employs the Back-Propagation Artificial Neural Network (BP-ANN) to predict the boundary points. Using the well-trained BP-ANN, this method can estimate the maximum gliding distances with only the initial altitude and radial directions. Simulations are conducted to analyze these two methods. Compared with conventional methods using point-mass flight dynamics, Method 1 considers more flight constraints, and the gliding footprint area is reduced by 20.79%. These results are relatively conservative and can improve the safety threshold of forced landing sites. Method 2 can estimate the gliding footprints (encircled by the boundary points under the entire operational altitude and full radial direction) in real time, which reserves more response and action time for aircraft forced landing.