基于时变量测方差的多传感器多目标优化分配算法  

Sensor Assignment Method Based on Time-varying Measurement Variance for Tracking Multi-targets

作　　者：方余瑜 左燕[1] 谷雨[1] 薛安克[1] 

机构地区：[1]杭州电子科技大学信息与控制研究所,杭州310018

出　　处：《火力与指挥控制》2015年第4期6-9,共4页Fire Control & Command Control

基　　金：国家自然科学基金(61004119;61174024);国家"九七三"计划基金资助项目(2012CB821204)

摘　　要：在目标跟踪过程中,目标的动态模型通常在笛卡尔坐标系中,而量测是在极/球坐标系中得到的。在基于卡尔曼滤波及其一些改进算法中,设定量测方差固定不变,可能导致滤波发散。为此提出了一种基于时变量测方差的多传感器资源管理算法。该算法通过统计方法求出转换测量值误差的均值和方差,利用转化卡尔曼滤波算法估计误差协方差,基于协方差的效能函数进行多传感器多目标分配。仿真结果显示该算法在目标跟踪过程中满足跟踪精度要求,并实现传感器资源的充分的利用。The target states are usually in Cartesian coordinates while the measurement values are in polar coordinates in target tracking system. The measurement variance is usually assumed fixed,which may lead to filter divergence in the algorithm based on kalman filter and its extension. Hence,an optimal sensor assignment method based on time varying measurement variance for tracking multi-targets is presented. It gets the time-varying measurement variance by converted measuring coordinates and estimates the covariance with the converted measurement kalman filtering algorithm. The sensor assignment for tracking multi-target can be implemented with the cost function of the covariance. Finally,the simulation shows that the algorithm can meet the tracking accuracy requirements while utilize the sensor resources fully.

关 键 词：目标跟踪 测量方差 传感器管理 多传感器多目标分配 转化卡尔曼滤波算法 

分 类 号：TN953[电子电信—信号与信息处理]