Neural-network midcourse guidance with consideration of the head-on attack condition

The objective of this study is to enhance neural-network guidance to consider the impact condition. Missile impact angle error, a measure of the degree to which the missile is not steering for a head-on attack, can have a significant influence on the final miss distance. Midcourse guidance using neural networks is employed to reduce the deviation angle from head-on effectively in the three-dimensional space. In addition, a coordinate transformation is introduced to simplify the three-dimensional guidance law and reduce training data for the neural network. Computational results show that the current neural-network guidance law with the coordinate transformation can be used to reduce the impact angle errors.