Real-time neural-network midcourse guidance

The approximation capability of artificial neural networks has been applied to the midcourse guidance problem to overcome the difficulty of deriving a n on-board guidance algorithm based on optimal control theory. This approac h is to train a neural network to approximate the optimal guidance law in f eedback form using the optimal trajectories computed in advance, Then the t rained network is suitable for real-Lime implementation as well as generati ng suboptimal commands. In this paper, the advancement of the neural-networ k approach to the current level from the design procedure to the three-dime nsional flight is described. (C) 2001 Published by Elsevier Science Ltd.