SRCKF-GM-PHD多目标跟踪算法的优化  

Optimization of SRCKF-GM-PHD Multi-target Tracking Algorithm

作　　者：齐美彬[1] 伍坤江 赵谦 徐晋[2] QI Meibin;WU Kunjiang;ZHAO Qian;XU Jin(School of Computer and Information,Hefei University of Technology,Hefei Anhui 230009,China;The 38th Research Institute of CETC,Hefei Anhui 230088,China)

机构地区：[1]合肥工业大学计算机与信息学院,安徽合肥230009 [2]中国电子科技集团公司第三十八研究所,安徽合肥230088

出　　处：《现代雷达》2023年第4期31-39,共9页Modern Radar

基　　金：国家自然科学基金资助项目(61771180)。

摘　　要：针对平方根容积卡尔曼滤波高斯混合概率假设密度(SRCKF-GM-PHD)算法在高杂波条件下对非线性目标跟踪能力弱的问题,文中首先融入改进灰狼算法,实时调节过程噪声Q和量测噪声R。其次,结合改进的渐消因子思想,实时调整SRCKF-GM-PHD算法中的增益矩阵,提高目标的跟踪精度。此外,为避免算法中止,文中提出动态权重调整策略的改进措施,调整算法中的实际输出残差序列的协方差。最终,形成了融合改进灰狼算法和改进渐消因子的SRCKF-GM-PHD算法。仿真分析对比了四种算法的性能,表明了所提算法在跟踪精度方面的有效性和优越性。Aiming at the problem that the square root cubature Kalman filter Gaussian mixture probability hypothesis density(SRCKF-GM-PHD)algorithm is not strong in tracking non-linear targets under the condition of the high clutter,in this paper firstly the improved grey wolf optimizer is infused to adjust the process noise Q and measurement noise R in real time.Secondly,the gain matrix of SRCKF-GM-PHD algorithm is adjusted to enhance the tracking accuracy of the target with the idea of improved fading factors.In addition,in order to avoid discontinuing the algorithm,an improved measure of dynamic weight adjustment strategy is proposed to adjust the covariance of the actual output residual sequence in the algorithm.Finally,a novel SRCKF-GM-PHD algorithm combining the improved grey wolf optimizer and the improved fading factors is formed.The performances of the four algorithms are compared by simulation analysis,and the validity and superiority of the proposed algorithm in tracking accuracy are demonstrated.

关 键 词：平方根容积卡尔曼 高斯混合概率假设 灰狼算法 渐消因子 目标跟踪 

分 类 号：TN957.52[电子电信—信号与信息处理]