基于IMMEPF的多普勒盲区目标异类多传感器联合跟踪

Heterogeneous Multi-Sensor Joint Tracking of Target Hidden in Blind Doppler Zone Based on IMMEPF

对于机载脉冲多普勒雷达,多普勒盲区是不可避免的。为解决多普勒盲区内机动目标跟踪问题,提出了基于扩展卡尔曼粒子滤波(IMMEPF)的雷达和ESM联合跟踪算法。该算法融合了交互式多模型(IMM)、粒子滤波(PF)和扩展卡尔曼滤波(EKF)的优势,采用多模型结构以匹配目标的运动模型。粒子滤波能处理非线性、非高斯问题,而采用EKF产生粒子,由于考虑了当前观测值,使得粒子的分布更接近后验概率密度分布,克服粒子的退化现象,从而提高估计精度。仿真结果表明,给出的算法能够显著提高对落入多普勒盲区内的目标点迹的跟踪精度。

For airborne pulse Doppler （PD） radar, the blind Doppler zone （BDZ） is inevitable. To solve the problem of tracking maneuvering targets in BDZ, a joint tracking algorithm of radar and ElectronicSupport Measure （ESM） based on extended Kalman particle filter （IMMEPF） was presented, which combined the advantages of interactive multiple model （EMM）, particle filter （PF） and extended Kalmanfilter （EKF）, and adopted multi-model structure to match the target＇s movement model. Particle filter was capable of handling the non-linear, non-Gaussian problem, and EKF was used to generate particle. Theparticle distribution was closer to a posterior probability density distribution due to its consideration of current measurements, which overcame the particle degradation phenomenon, thus improving the estimationaccuracy. Simulation resuhs showed that the given algorithm could significantly improve the tracking accuracy of targets hidden in BDZ.