Nonlinear Flight Control Using Neural Networks

Byoung S. Kim, Anthony Calise

Journal of Guidance Control and Dynamics · 1997 · 591 citations · 15 references

Concepts

TL;DR

Neural networks are emphasized for use in flight control architectures that employ feedback linearization of aircraft dynamics. The study develops a direct adaptive tracking control architecture that uses neural networks capable of online learning to compensate for inversion errors in aircraft dynamics. Neural networks model the nonlinear inverse transformation for feedback linearization, are first trained offline on a nominal model, and then updated online with a stable weight‑adjustment rule to compensate for modeling errors. The adaptation algorithm guarantees bounded loop signals and convergence of neural‑network weights, and simulations on an F‑18 model demonstrate its effective performance.

Abstract

The theoretical developmentofa direct adaptivetracking controlarchitectureusingneuralnetworks ispresented. Emphasis is placed on utilization of neural networks in a  ight control architecture based on feedback linearization of the aircraft dynamics. Neural networks are used to represent the nonlinear inverse transformation needed for feedback linearization. Neural networks may be Ž rst trained off line using a nominalmathematicalmodel, which provides an approximate inversion that can accommodate the total  ight envelope. Neural networks capable of on-line learning are required to compensate for inversion error, which may arise from imperfect modeling, approximate inversion, or sudden changes in aircraft dynamics. A stable weights adjustment rule for the on-line neural network is derived. Under mild assumptions on the nonlinearities representing the inversion error, the adaptation algorithm ensures that all of the signals in the loop are uniformly bounded and that the weights of the on-line neural network tend to constant values. Simulation results for an F-18 aircraft model are presented to illustrate the performance of the on-line neural network based adaptation algorithm.

References

15