防空导弹引信和导引头探测信息的融合优化研究  

Fusion and Optimality of Ground-to-air Missile Fuze and Seeker Target Detection Information

作　　者：韩俊杰[1] 李为民[1] 黄树采[1] 

机构地区：[1]空军工程大学导弹学院,陕西三原713800

出　　处：《宇航学报》2006年第2期210-212,296,共4页Journal of Astronautics

基　　金：高等学校骨干教师资助计划基金项目(GG-110-90039-1004)

摘　　要：分析防空导弹引信和导引头探测目标信息的互补性和信息融合优化的可行性,提出了前馈神经网络MLSP的优化步长OBP学习算法。采用最优化的原理优化步长的选取很好的解决了传统BP算法收敛速度慢和产生振荡的缺点,构造了一个用于防空导弹引信和导引头探测目标信息的融合优化模型,并以某型防空导弹武器系统的引信和导引头分系统为例,仿真结果说明了该模型的有效性和可用性。Based on the targect detection information complementarity of ground-to-air missile fuze and seeker, the feasibility of their fusion and optimality is analyzed. An optimal learning rate OBP algorithm of the feed forward neural network is proposed. The problems of convergence rates and oscillations is corrected in the favor of optimality method. The fusion and optimality model used on target detection information of the ground-to-air missile fuze and seeker is structured. In the finial, by an experiment dealing with certain fuze and seeker, the usability and effectiveness of the model is testd.

关 键 词：防空导弹 引信 导引头 前馈神经网络 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]