摘要
针对高斯混合概率假设密度(GMPHD)滤波算法中的机动目标跟踪问题,提出了一种改进的最佳拟合高斯(BFG)与GMPHD结合的BFG-GMPHD算法。算法对BFG近似方式做出改进,通过匹配状态转移的均值和协方差矩阵来近似多个目标动态模型中的状态转移矩阵和过程噪声的协方差矩阵,实现了滤波器与不同动态模型的匹配;在对BFG分布进行递推时,引入了模型概率更新过程,解决了BFG仅依赖于先验信息的问题。仿真实验表明:改进后的算法能滤除传感器数据中的杂波干扰,有效匹配目标运动模型的变化,更加准确地估计出目标的数目和状态,提高了跟踪的性能。
In order to track maneuvering target with Gaussian mixture probability hypothesis density（ GMPHD）filtering algorithm,a new algorithm combines improved best fitting Gaussian（ BFG） with GMPHD,that is BFGGMPHD algorithm,is proposed. The approximation method is improved in the proposed algorithm which approximates the state transition matrix and process noise covariance matrix of target kinematic model by matching the transition mean and covariance matrix. The model probability update process is introduced into the recursion of BFG to solve the problem that the recursion of BFG is only determined by priori information. Simulation experiments show that the improved BFG-GMPHD algorithm can filter out the clutter in sensor data,effectively match change of target moving model,accurately estimate number and state of targets and improve the performance of tracking.
出处
《传感器与微系统》
CSCD
2016年第7期136-139,146,共5页
Transducer and Microsystem Technologies
关键词
高斯混合概率假设密度
机动目标
改进最佳拟合高斯
模型概率更新
Gaussian mixture probability hypothesis density（GMPHD）
maneuvering targets
improved best fitting Gaussian（BFG）
model probability update
