Time cooperative guidance law based on neural network

Abstract

In response to the issue of coordinated guidance for vehicle groups, a neural network-based time-to-go prediction method has been designed. The influence of the motion state of both the vehicle and the target on the prediction results has been considered in this method. Compared with the classical method, the prediction error of the moving target is obviously reduced. A guidance law with a time synergy coefficient is proposed. This method adjusts the flight trajectory of the vehicles by the time-to-go error, ensuring that the cluster of vehicle arrive at the target simultaneously. According to the simulation results, the prediction error of the neural network is less than 0.4 s, and the maximum coordination error is 0.2 s, which achieves the effect of cooperative interception.