A Neighboring Optimal Adaptive Critic for Missile Guidance

Author

J. Dalton
S. N. Balakrishnan, Missouri University of Science and TechnologyFollow

Abstract

We present a neural network approach to missile guidance which is based on the notion of an adaptive critic. This approach is derived from the use of both a nominal solution of a linear optimal guidance law and neighboring optimal control law. No assumptions about target maneuver dynamics are made during neural network training. We discuss neuro-control training issues, and the neural network control system results are compared with those obtained from an optimal control formulation. Numerical results from the simulations of the neuro-controller under reference conditions and under perturbations due to target maneuvers are presented. We also demonstrate the transfer of control knowledge from the critic network to the controller network while the simulated missile is in flight.

Recommended Citation

J. Dalton and S. N. Balakrishnan, "A Neighboring Optimal Adaptive Critic for Missile Guidance," Mathematical and Computer Modelling, Elsevier, Jan 1996.

The definitive version is available at https://doi.org/10.1016/0895-7177(95)00226-X

Department(s)

Mechanical and Aerospace Engineering

International Standard Serial Number (ISSN)

0895-7177

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 1996 Elsevier, All rights reserved.

Publication Date

01 Jan 1996