| 摘要: |
| 目标跟踪广泛应用于要地防御、防空反导和无人驾驶等军事和民用领域。相比单目标跟踪，多目标跟踪往往涉及到未知数目的多个目标以及杂波、漏检等复杂情况，所面临的一个技术难点是数据关联，包括量测与目标航迹之间的关联以及不同传感器之间航迹关联等。本文梳理了多目标跟踪应用中数据关联主要解决思路，首次将经典数据关联方法分为确定性数据关联和概率性数据关联，前者包括最近邻、全局最近邻和多假设跟踪，后者包括概率数据关联、联合概率数据关联和概率多假设跟踪，系统阐述和对比了这些典型算法及其扩展方法的基本原理、各自适用条件和优缺点，揭露它们之间的关联性，并指出近年来智能学习、优化算法等也为数据关联问题提供了新的解决思路。特别是在视觉跟踪领域，关联跟踪与目标特征学习、场景感知密切结合，给数据关联带来了新的挑战。本文对上述各类方法和思路进行了分析和总结，并展望了未来发展趋势。 |
| 关键词: 多目标跟踪 数据关联 确定性关联 概率性关联 |
| DOI： |
| 分类号:TN953+.6 |
| 基金项目:国家自然科学基金(62071389);陕西省自然科学基础研究计划(2023JC-XJ-22);中央高校基本科研业务费专项资金 |
|
| A Survey on Data Association Approaches to Multi-Target Tracking |
|
|
|
|
| Abstract: |
| Target tracking is widely used in military and civilian fields, such as attack and defense in important areas, air and missile defense, and unmanned driving. Compared with single target tracking, multi-target tracking often involves an unknown number of targets as well as clutter, missing detection and other complicated situations. One of the core technical difficulties is the data association, including that between the measurement and targets, and that between target tracks from different sensors. This survey reviews the main solutions of the data association (DA) problem involved in multi-target tracking, classifying classic approaches into deterministic and probabilistic DA for the first time. The former includes the nearest neighbor, global nearest neighbor, and multi-hypothesis tracking, while the latter includes probabilistic DA, joint probabilistic DA, and probabilistic multi-hypothesis tracking. Systematic elaboration and comparison of the basic principles, respective applicability, advantages and disadvantages of these classic algorithms and their extensions are provided, exposing their interrelationship. It is further pointed out that the intelligent learning and optimization algorithms have also provided new solutions for DA problems in recent years. Especially in the context of visual tracking, the combination of DA tracking and target feature learning, scene awareness causes new challenges. This survey analyzes and summarizes these various DA methods and ideas, and looks to the future development trend. |
| Key words: multi-target tracking data association deterministic association probabilistic association |