基于高斯粒子JPDA滤波的多目标跟踪算法  

Gaussian Particle JPDA Filter Based Multi-target Tracking

作　　者：张俊根[1] 姬红兵[1] 蔡绍晓[1] 

机构地区：[1]西安电子科技大学电子工程学院,西安710071

出　　处：《电子与信息学报》2010年第11期2686-2690,共5页Journal of Electronics & Information Technology

基　　金：国家自然科学基金(60871074)资助课题

摘　　要：在多目标跟踪中,由于观测的不确定性带来数据关联问题,并且,多目标状态空间尺寸的增长带来了维数增大问题,该文提出了一种新的高斯粒子联合概率数据关联滤波算法(GP-JPDAF),在JPDA框架中引入高斯粒子滤波(GPF)的思想,通过高斯粒子而不是高斯量,来近似目标与观测的边缘关联概率,利用GPF计算目标状态的预测及更新分布。将其应用于被动多传感器多目标跟踪,仿真结果表明该算法比MC-JPDAF具有更好的跟踪性能。In multi-target tracking,aiming at the data association problem that arises due to indistinguishable measurements in the presence of clutter,and the curse of dimensionality that arises due to the increased size of the state-space associated with multiple targets,a novel algorithm based on Gaussian Particle Joint Probabilistic Data Association Filter（GP-JPDAF） is proposed,which introduces Gaussian Particle Filtering（GPF） concept to the JPDA framework.For each of the targets,the marginal association probabilities are approximated with Gaussian particles rather than Gaussians in the JPDAF.Moreover,GPF is utilized for approximating the prediction and update distributions.Finally,the proposed method is applied to passive multi-sensor multi-target tracking.Simulation results show that the method can obtain better tracking performance than Monte Carlo JPDAF（MC-JPDAF）.

关 键 词：多目标跟踪 联合概率数据关联 高斯粒子滤波 被动多传感器 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]