基于模糊混合退火分布的多目标高斯混合粒子PHD滤波算法  

Gaussian Mixture Particle Probability Hypothesis Density Filter Based on Fuzzy Hybrid Annealed Distribution in Multi-target Tracking

作　　者：冉星浩 陶建锋[1] 贺思三[1] RAN Xinghao;TAO Jianfeng;HE Sisan(Air and Missile Defense College,Air Force Engineering University,Xi’an 710051,China)

机构地区：[1]空军工程大学防空反导学院

出　　处：《弹箭与制导学报》2019年第3期130-134,139,共6页Journal of Projectiles,Rockets,Missiles and Guidance

基　　金：国家自然科学基金(61501495)资助

摘　　要：针对杂波环境下高斯混合粒子PHD跟踪精度低,滤波发散的问题,提出基于模糊混合退火分布的高斯混合粒子PHD。所提算法在传统的高斯混合粒子PHD滤波的基础上,采取状态变量分解和引入退火参数产生建议分布函数,同时采用模糊推理系统产生最优的退火系数,有利于提高粒子滤波的稳定性和精度,然后对PHD进行更新。仿真结果表明:该算法能在杂波环境下有效跟踪多个目标,与高斯混合粒子PHD滤波相比,状态估计更加接近真实值,大大提高了跟踪精度和系统稳定性。In order to overcome lower estimating accuracy and filtering divergence of traditional GMP-PHD algorithm in clutter environment,a new improved GMP-PHD is proposed in this paper,which based on fuzzy hybrid annealed distribution.Compared with the traditional GMP-PHD,the algorithm adopts the state variable decomposition and the introduction of annealing parameters to generate the proposed distribution fuction,and the optimal annealing coefficient is generated by the fuzzy inference system,which could improve the stability and accuracy of particle filter,and then to update a PHD.The simulation show that the improved algorithm can effectively track multiple targets in clutter environment,compared with GMP-PHD filter,closer to the true value,improves the tracking accuracy and system stability.

关 键 词：概率假设密度滤波 混合退火分布 多目标跟踪 高斯混合粒子PHD 

分 类 号：TN713[电子电信—电路与系统] TN953