基于遗传规划的空空导弹攻击区拟合  

Air-to-air Missile Launch Envelops Fitting Based on Genetic Programming

作　　者：李枭扬 周德云[1] 冯琦[1] 张华涛 

机构地区：[1]西北工业大学电子信息学院,西安710072 [2]中国兵器工业第203研究所,西安710065

出　　处：《弹箭与制导学报》2015年第3期16-18,22,共4页Journal of Projectiles,Rockets,Missiles and Guidance

基　　金：航空科学基金(No.20115553021)资助

摘　　要：针对传统空空导弹攻击区多项式拟合方法拟合模型固定,拟合精度较差等问题,提出一种基于遗传规划的空空导弹攻击区拟合方法。首先建立遗传规划空空导弹攻击区拟合模型,然后利用遗传规划算法对攻击区进行拟合,得到攻击区拟合多项式模型,最后通过非线性回归的方法对拟合得到的多项式模型进行回归分析,修正模型系数,得到最终的空空导弹攻击区拟合多项式。仿真实验证明了该算法比传统多项式拟合算法拥有更高的拟合精度。For the problem that fitting model is fixed and precision is low of traditional air-to-air missile launch envelops fitting method, an air-to-air missile launch envelops fitting method based on genetic programming was proposed. Firstly, an air-to-air missile launch envelops fitting model based on genetic programming was established. Secondly, the air-to-air missile launch envelops was fitted by genetic program- ming algorithm so as to get polynomial fitting model. Finally, regression analysis on the polynomial model was performed by nonlinear re- gression method in order to update the model coefficient and get the final air-to-air missile launch envelops fitting polynomial. The simula- tion results show that this algorithm is higher in precision than traditional polynomial fitting algorithm.

关 键 词：空空导弹攻击区 遗传规划 曲线拟合 非线性回归 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]