Abstract:

Since the traditional threat assessment can not adjust the weights dynamically according to the changing of threat factors during beyondvisualrange air combat, the back propagation (BP) neural network is introduced. Considering the subjective factors and objective factors, the structure entropy weight method is used to confirm the weights of each threat index and the weights are supplied to BP neural network training. The dynamic weights calculating method is proposed based on the improved glowworm swarm optimization (IGSO)algorithm and the BP neural network. In IGSO-BP, IGSO is employed to simultaneously optimize the initial weights and thresholds of the BP neural network. The optimized BP network can calculate the weights in different situations better. On the circumstance of assessing the threat degree during multiunmanned combat air vehicle (UCAV) cooperation combat, simulation results based on the structure entropy weight method and the IGSO-BP algorithm indicate that, the method can assess the weights of threat indexes effectively and can possess better reliability and veracity than conventional methods.