改进IMM的两飞行体无源定位跟踪算法  

Improved Interacting Multiple Model Filter Algorithm of Passive Localization Between Two Aircraft Bodies

作　　者：高宪军[1] 霍长庚[1] 谈欣荣[1] 

机构地区：[1]空军航空大学科研部,吉林长春130022

出　　处：《现代防御技术》2014年第2期61-66,共6页Modern Defence Technology

摘　　要：IMMPF算法巨大的计算量影响跟踪的实时性。针对这一问题,在基于相位差变化率的两飞行体无源定位问题的基础上提出了一种改进的交互式多模型滤波算法(IMMK-UKF-PF),利用不同的模型匹配不同类型的滤波器,充分发挥了粒子滤波和无迹卡尔曼滤波以及卡尔曼滤波各自的优点。仿真结果表明,该算法大大提高了计算效率,减少了跟踪定位所用时间,同时具有良好的跟踪性能和较强的鲁棒性。Interacting multiple model particle filter＇s （IMMPF） heavy computational load affects the real-time of tracking. To solve the problem, an improved interacting multiple model filter （IMMK-UKF- PF）, based on the positioning method of phase change rate between two aircraft bodies, is proposed. The algorithm makes use of different models to match different filters and gives full play to particle filter and unscented Kalman filter （UKF） and Kalman filter （KF）. The simulation results show that the new filte- ring method improves computational efficiency greatly and decreases the time of positioning and tracking. At the same time, the new algorithm has better tracking performance and strong robustness.

关 键 词：相位差变化率 无源定位 交互式多模型 粒子滤波 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] TN958.97[自动化与计算机技术—计算机科学与技术]