Aircraft flight path reconstruction with nonlinear adaptive filters

The objective of the present paper is to develop a modified recursive prediction error adaptive filter based on the maximum likelihood concept for online nonlinear aircraft flight path reconstruction problems. The new algorithm revises the general recursive prediction error methods by modifying the prediction error vector calculation, which was basically independent of the parameter estimator in all well-known prediction error methods, and considers the mutual coupling between the parameter estimator and the prediction error estimator in order to improve the asymptotical behaviour and accuracy of the nonlinear adaptive filter especially during the transient period. Numerical results are presented for a nonlinear aircraft model in a symmetrical flight condition. System states as well as parameters including measurement noise standard deviations of the flight test instrumentation system are estimated by different algorithms. The result gives a clear picture that the modified recursive maximum likelihood adaptive filter, as developed in this paper, produces more robust and accurate estimates as compared to the general recursive prediction error methods