神经网络在直升机对地攻击中的应用研究

Application Study of Neural Network in Ground-attack by Helicopter

应用神经网络方法研究了直升机对地射击的一些技术问题。文中首先介绍了直升机对地攻击的火控系统原理;其次,对复杂又不稳定的直升机飞行运动模型设计了模糊神经网络控制器;然后,设计了综合火/飞控制系统,并对所设计的整个攻击系统进行了仿真,通过仿真证明满足系统设计要求。在神经网络设计中,采用了变尺度优化算法,提高了算法的速度和精度,直升机对地攻击的实时应用提供了重要的参数价值。

Some technology issues about ground-attack by helicopter are studied using neural network.Firstly the principle of fire control system about ground-attack by helicopter is described;Secondly,because the armed helicopter model is complex and less stability,A fuzzy neural network controller (FNNC) is designed;Lastly,the whole process of designed system is simulated and results meet the design requirements.Optimizing algorithm of MDFP is used in the design of neural network,Which improve velocity and precision of algorithm.Reference value for real-time application in ground attack by helicopter is provided.