Missile Longitudinal Autopilot Design Using a New Model-Following Robust Neuro-Adaptive Controller


Neural networks have become popular in recent years in designing robust and adaptive controllers for dynamic systems because of their universal function approximation property. This paper develops solutions for a model-following controller which is applicable for general systems of known order with different kinds of uncertainties, such as missing functions in the differential equations (unmodeled dynamics) and/or parameter variations. The controller design is carried out in two steps: (i) synthesis of a set of neural networks that capture the unmodeled dynamics and parametric uncertainties of the plant on-line (ii) computation of a controller that drives the states of the plant to that of a desired nominal model. The neural network weight update rule has been derived using the Lyapunov theory that guarantees both stability of the error dynamics as well as boundedness of the weights. The novelty of the new technique is that it can be used in conjunction with any known control design technique for the nominal model and makes the plant behavior robust to unmodeled dynamics and parametric uncertainties. This powerful approach has been applied in designing a missile longitudinal autopilot controller and numerical results from simulation studies have been presented here.


Mechanical and Aerospace Engineering


National Science Foundation (U.S.)

Keywords and Phrases

Adaptive Control Systems; Automatic Pilots; Longitudinal Control; Missile Control; Neural Networks; Robust Stabilization

Document Type

Article - Conference proceedings

Document Version


File Type





© 2004 American Institute of Aeronautics and Astronautics (AIAA), All rights reserved.