一种姿态角辅助的IMMPF算法  

Pose-angle Aided IMMPF Algorithm

作　　者：周云锋[1] 单甘霖[1] 

机构地区：[1]军械工程学院光学与电子工程系,石家庄050000

出　　处：《系统仿真学报》2010年第7期1609-1613,共5页Journal of System Simulation

摘　　要：交互式多模型粒子滤波(IMMPF)算法把粒子滤波(PF,Particle Filter)引入交互式多模型估计(IMM)算法,从而能够有效地解决非线性、非高斯机动目标跟踪问题。首次把姿态角信息引入到针对空中机动目标跟踪的IMMPF算法中,通过姿态角测量与当前运动模式的模糊关联来辨识目标的机动模式,然后把辨识结果与IMMPF算法的后验粒子权值相融合,以提高算法本身的模型分辨能力;对机动目标跟踪的仿真实验表明,该方法能够有效地改善原跟踪算法的跟踪精度和稳定性。Interacting Multiple Model Particle Filtering（IMMPF） algorithm brings Particle Filter（PF） into the Interacting Multiple Model（IMM） estimation algorithm,to effectively solve the tracking problem of Non-Gaussian and Non-linear maneuvering target.The pose angle information was firstly brought into the IMMPF algorithm,and the maneuvering mode of target was identified through the fuzzy association between the pose angle and the current motion mode of target;and then the association result was used to fuse with model probability of IMM to enhance its model resolving power.A simulation of maneuvering target tracking shows that the method can effectively improve the tracking accuracy and stability of the original IMMPF algorithm.

关 键 词：交互式多模型 粒子滤波 目标跟踪 仿真 姿态角 模糊关联 

分 类 号：TN953[电子电信—信号与信息处理]