基于随机有限集理论的多目标跟踪方法若干问题研究

项目名称： 基于随机有限集理论的多目标跟踪方法若干问题研究

项目编号： No.61201118

项目类型： 青年科学基金项目

立项/批准年度： 2013

项目学科： 电子学与信息系统

项目作者： 陈金广

作者单位： 西安工程大学

项目金额： 24万元

中文摘要： 多目标跟踪技术在航空航天、军事防卫、机器人等领域具有重要的应用价值。基于随机有限集理论的多目标跟踪方法，能够避免数据关联步骤的困扰，能够较好地解决复杂环境中目标数目未知且随时间变化的多目标跟踪问题，已成为近年来国际上研究的热点。 项目针对基于随机有限集理论的多目标跟踪方法中的若干关键问题展开研究，致力于提高多目标跟踪的有效性。首先，研究复杂条件下更准确的目标状态提取方法。在跟踪过程中对目标进行标定，形成多目标航迹，利用已经形成的航迹信息重新设计目标状态提取算法，以获得更有效的目标状态估计方法。其次，对多目标跟踪场景进行研究，寻找目标跟踪系统中惯有的约束关系，并在滤波过程中加以合理利用，进一步提高滤波的精度，从而提高目标跟踪的精度。此外，考虑到信号采集过程中存在同一目标持续多步漏检的情况，研究一般意义下的航迹关联技术。 通过本项目研究，可以丰富和补充多目标跟踪方法的实现途径和理论体系。

中文关键词： 随机有限集；概率假设密度滤波；目标跟踪；状态估计；信息融合

英文摘要： Multi-target tracking techniques have important application value in many areas, such as aerospace, military defense, and robotics. Multi-target tracking methods based on random finite set theory can avoid problems appeared in the progress of data association. It can also deal with multi-target tracking problems in the complex environment with target number unknown and target number varying with time. It has become an international focus of research in recent years. The proposal aims at some key problems about multi-target tracking methods based on the random finite set theory. It devodes ourselves to the studies of promoting the effectiveness of multi-target tracking. Firstly, we will study new methods of more accurate target state extraction in complex situations. Targets are labeled in the process of tracking, and then multi-target tracks are formed. More effective algorithms for target state extraction can be obtained by using the former track information. Subsequently, we will study some multi-target tracking scenarios in real application to seek general constraints of tracking system. More accurate results will be obtained as the constraints are applied properly in the filter. Furthermore, considering the case of continuous multi-step miss-detection on the same target in the progress of signal collection,

英文关键词： random finite set；probability hypothesis density filter；target tracking；state estimation；information fusion