多传感器稳健融合跟踪算法  

A Robust Fusion Algorithm Based on Multi-Sensor Tracking

作　　者：胡士强[1] 敬忠良[1] 胡洪涛[1] 田宏伟[1] 李建勋[1] 

机构地区：[1]上海交通大学自动化系

出　　处：《上海交通大学学报》2005年第4期509-512,共4页Journal of Shanghai Jiaotong University

基　　金：国家自然科学基金项目(60375008;60304007);河北省博士科学基金项目(B2004510);航天支撑技术基金项目(JD04)

摘　　要：讨论了多传感器融合跟踪的稳健性算法.针对集中式多传感器的融合跟踪结构,采用统计方法和随机逼近方法分析了传感器最优权的选取原则,得出了传感器融合对公共测量噪声没有影响的结果.依据最优选取原则给出了两种自适应融合跟踪算法,算法能在线适应传感器性能的变化,并使融合方差最小.采用典型航路进行了算法仿真,结果验证了理论分析的合理性和工程应用有效性.The robustness of fusion algorithm based on multi-sensor tracking was discussed. Aiming at the centralized structure of multi-sensor tracking system, the optimal weighting coefficients are obtained using statistical estimation and stochastic approximation, from which it is concluded that sensor fusion has no effect on common measurements noise. Two algorithms were presented, which are able to adapt themselves to the changes of sensor's measurements noise, and the estimation error is of least mean square. The simulation results indicate that the fusion algorithm has good robustness and is very potential in practical engineering system.

关 键 词：数据融合 稳健跟踪 模糊推理 多传感器 

分 类 号：TP273.3[自动化与计算机技术—检测技术与自动化装置] V248.1[自动化与计算机技术—控制科学与工程]