论文部分内容阅读
针对高超声速巡航导弹的高空巡航飞行段,采用虚拟目标的定义方法,结合微分几何知识以及运动学方程建立了导弹与目标的相对运动模型,并在此基础上视其与虚拟目标之间的运动为单目标微分对策问题。利用哈密尔顿函数求解方法推导了开环微分对策中制导律,同时建立了一种新的闭环微分对策中制导律结构图,并对反向传播神经网络的训练样本进行了设计。通过反向传播神经网络的函数逼近功能实现了高超声速巡航导弹微分对策中制导律的智能化。仿真验证表明了其有效性。
Aiming at the high altitude cruise flight segment of hypersonic cruise missile, the definition method of virtual target is adopted, and the relative motion model between missile and target is established based on differential geometry knowledge and kinematics equation. Based on this, the motion between the missile and the target is considered For single-objective differential game problem. The guidance law of the open-loop differential game was deduced by solving the Hamiltonian function. At the same time, a new guidance law structure diagram of closed-loop differential game was established and the training samples of the backpropagation neural network were designed. Through the function approximation function of backpropagation neural network, the intelligence of guidance law in hypersonic cruise missile differential game is realized. Simulation results show its effectiveness.