针对导引头系统内部产生干扰力矩引起的隔离度问题,分析了影响干扰力矩的主要因素,通过神经网络对跟踪环路进行建模,对影响因素反复进行学习和训练,产生一个与干扰力矩相近的力矩对跟踪控制回路进行前馈补偿,使系统输出的干扰力矩减小,同时提升了导弹的性能。仿真实验表明,将人工神经网络引入导引头跟踪控制回路中,可以有效地解决干扰力矩带来的隔离度问题,减小导弹的脱靶量,使导弹的命中精度提高。

In view of the isolation problems caused by the seeker system internal disturbance torque, this paper analyzed the main factors influencing the disturbing torque. A model for tracking loop was made by the neural network, and iterative learning and training of the influencing factors was made to produce a similar disturbance torque to give feedforward compensation of tracking control loop. It reduced the interference of the system output disturbance torque, while increased the performance of the missile. Simulation results showed that the artificial neural network was effectively introduced into seeker tracking control circuit to solve the problem of isolation caused by disturbance torque, reduce the miss distance and increase the accuracy of missiles.