Premium
Synthesis of optimal feedback guidance law for homing missiles using neural networks
Author(s) -
Rahbar N.,
Bahrami M.
Publication year - 2000
Publication title -
optimal control applications and methods
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.458
H-Index - 44
eISSN - 1099-1514
pISSN - 0143-2087
DOI - 10.1002/1099-1514(200005/06)21:3<137::aid-oca668>3.0.co;2-e
Subject(s) - homing (biology) , missile , artificial neural network , missile guidance , proportional navigation , control theory (sociology) , computer science , terminal guidance , guidance system , line of sight , sight , law , artificial intelligence , engineering , control (management) , aerospace engineering , ecology , physics , astronomy , political science , biology
Most existing missiles are guided by proportional navigation guidance (PNG) law, but PNG is a particular case for LQ guidance rule with two main assumptions of small line‐of‐sight angles and negligible acceleration along the line‐of‐sight. However, most missile engagements exceed these limits because of high tangential and normal accelerations. Unfortunately, it is not possible to determine the feedback guidance law for non‐linear systems such as homing missiles in real‐time. We use artificial neural networks to synthesize feedback laws for homing missiles with non‐linear state equations. We first obtain an open‐loop optimal numerical solution for non‐linear state equations and then use these data to train a feed‐forward multilayer neural network in an off‐line session. The network is then used effectively in a real‐time for feedback guidance method. Simulation results show that this neural networks guidance method can efficiently produce an optimal feedback law in spite of relatively simple network architecture. Copyright © 2000 John Wiley & Sons, Ltd.