基于实神经网络的空间测向定位算法  

A Three-dimensional Direction Location Algorithm Based on Real Neural Networks

作　　者：刘李楠[1,2] 赵晓萌[3] 

机构地区：[1]海军蚌埠士官学校,安徽蚌埠233012 [2]河海大学,南京211100 [3]安徽科技学院,安徽滁州233100

出　　处：《火力与指挥控制》2014年第6期48-51,共4页Fire Control & Command Control

基　　金：国家自然科学基金(11173010);安徽省级高校自然科学基金(KJ2012Z 056);安徽科技学院中青年科研基金资助项目(No.ZRC2011270)

摘　　要：目前三维空间测向定位多采用先在二维体系中计算然后转换为空间坐标的方法,正北偏差、投影变换的方位变形等对定位精度造成很大的影响。算法立足于三维空间,将实神经网络代数算法首次应用到三维空间测向定位,利用其非线性逼近能力强等特性求解多个测向平面形成的空间目标观测方程组。仿真实验表明,定位精度明显提高,在不同测向精度条件下定位误差接近CRLB,具有很强的实用性。At present the three-dimensional spatial location usually first calculate in two dimensional system,then convert to spatial coordinate. But north deviation azimuth deformation、projection transformation has a great effect on positioning precision. This algorithm is based on the three-dimensional space,the real algebra algorithm of neural networks is first applied to the three-dimensional space positioning,using the nonlinear approximation ability to solve equation set which formed with a plurality of direction finding plane. Simulation results show that this algorithm positioning accuracy is obviously improved,positioning error is closing to CRLB under the condition of different direction finding precision,so the algorithm is applicable to complex measurement environment.

关 键 词：实神经网络 三维空间 测向定位 

分 类 号：TN955[电子电信—信号与信息处理]