标签随机有限集下多机动目标跟踪快速实现  

Fast Implementation for Multiple Maneuvering Targets Tracking Based Labeled Random Finite Set

作　　者：刁丹丹[1,2] 房亮 DIAO Dan-dan;FANG Liang(The 27^(th)Research Institute of CETC,Zhengzhou 450047,China;School of Electronic Engineering,Xidian University,Xi'an 710071,China)

机构地区：[1]中国电子科技集团公司第二十七研究所,河南郑州450047 [2]西安电子科技大学电子工程学院,陕西西安710071

出　　处：《中国电子科学研究院学报》2023年第6期503-512,共10页Journal of China Academy of Electronics and Information Technology

摘　　要：针对传统δ-GLMB(δ-Generalized Labeled Multi-Bernoulli)滤波和δ-JGLMB(δ-Joint Generalized Labeled Multi-Bernoulli)滤波计算复杂度较高和实际场景中目标机动的问题,文中提出了IMM-δ-FGLMB(Interactive Multi-modelδ-GLMB)滤波和IMM-δ-FJGLMB(Interactive Multi-modelδ-JGLMB)滤波两种跟踪算法。在滤波新生阶段提出量测驱动Bernoulli成分筛选过程(Measurement Driven for Filtrate the Bernoulli Components,MDFBC),通过利用量测信息来对新生的Bernoulli成分进行粗筛,去除无贡献的新生Bernoulli成分,进而为算法减少计算量。在滤波的预测阶段,通过利用多个模型对机动目标的状态进行交互式预测,从而避免使用单个模型进行状态预测时出现的模型失配问题。实验验证了所提算法在多机动目标跟踪和计算效率方面的优良性能。Aiming at the problems of maneuvering target tracking and the high computational complexity of traditionalδ-Generalized labeled multi-Bernoulli(δ-GLMB)filter andδ-Joint generalized labeled multi-Bernoulli(δ-JGLMB)filter.This paper proposes two tracking algorithms,named interactive multi-modelδ-GLMB(IMM-δ-FGLMB)filter and interactive multi-modelδ-JGLMB(IMM-δ-FJGLMB)filter.In the newborn stage of filter,the measurement drive for filtrate the Bernoulli components(MDFBC)algorithm is proposed to remove the non-contributing newborn Bernoulli components,thereby reducing the amount of calculation for the algorithm.In the prediction stage,interactive multi-model algorithm are employed to predict the state of the maneuvering targets,thereby avoiding the“model mismatch”problem that occurs when using single model for state prediction.Numerical experiments verify the excellent performance of the proposed algorithms in terms of multi-maneuvering targets tracking and computational efficiency.

关 键 词：多机动目标追踪 δ-GLMB滤波 δ-JGLMB滤波 δ-FJGLMB滤波 交互式多模型算法 

分 类 号：TN954.6[电子电信—信号与信息处理]