基于抗差估计的三维测向无源定位算法  

A Robust Estimation Based 3D Passive Locating Algorithm

作　　者：莫成坤 陈树新[1] 吴昊[1] 张艺航 

机构地区：[1]空军工程大学信息与导航学院,西安710077

出　　处：《电光与控制》2015年第2期22-26,共5页Electronics Optics & Control

基　　金：国家自然科学基金(61071014)

摘　　要：针对方位角和俯仰角存在异常误差影响机载单站三维无源定位结果准确性的问题,提出一种鲁棒的递推最小二乘(RRLS)定位算法。在最小二乘(LS)解的基础上,利用M估计原理求出目标位置的鲁棒估计形式。为保证定位的实时性,推导了鲁棒的递推最小二乘算法,并选取IGGⅢ权函数作为等价权函数。理论分析表明,RRLS算法能够准确识别异常误差,通过降低异常观测数据的权值来减小其对定位准确性的影响。仿真结果表明,在两角存在异常误差时,RRLS算法具有良好的抗异常误差能力,保证了定位结果的有效性和快速收敛。To solve the problem of location inaccuracy caused by outliers in azimuth and elevation in single- observer airborne 3 D passive locating, a robust recursive least squares （RRLS） algorithm was proposed. On the basis of least square solution, the robust estimation of target location was obtained according to M-estimation principle. In order to ensure real-time locating, the robust recursive least squares algorithm was deduced, and the IGGm weight function was selected as equivalent weight function. Theoretical analysis indicated that the RR[S algorithm can identify outliers effectively, and reduce the influence of abnormal measurement data on locating result by decreasing the weights of them. Simulations show that when outliers exist in azimuth and elevating, RRLS algorithm has the good ability to resist the outliers, ensure the validity of the positioning result and fast convergence.

关 键 词：机载单站定位 无源定位 递推最小二乘估计 鲁棒估计 等价权函数 

分 类 号：V271.4[航空宇航科学与技术—飞行器设计] TP971[自动化与计算机技术]