VIRTUAL FLIGHT DATA RECORDER FOR COMMERCIAL AIRCRAFT

This paper presents preliminary results of the development of a virtua l flight data recorder (VFDR) for commercial airliners. Federal Aviati on Administration (FAA) regulations, currently being revised, mandate the recording of 11 dynamic parameters, not including the control surf ace deflections. The absence of these data can be critical for crash i nvestigation purposes. This paper proposed the introduction of a VFDR based on a neural network simulator (NNS) and a neural network reconst ructor (NNR). The NNS is trained, using flight data for the particular aircraft, to simulate any desired control surface deflections (or any other parameter of interest not recorded by the FDR), minimizing a co st function based on the differences between the available data from t he FDR and the output from the NNS. The VFDR scheme has been introduce d, tested, and validated with flight data from a Boeing 737-300 with a n FDR with extended recording capabilities showing accurate reconstruc tion of the control surface deflections' time histories. The VFDR can be considered a tool for crash investigations where control surface fa ilures are believed to be a factor.