{
  "paper_id": "5cae0969440d1c215f1c2016200676df",
  "paper_title": "People Tracking with Mobile Robots Using Sample-based Joint Probabilistic Data Association Filters",
  "paper_authors": [
    "Dirk Schulz",
    "Wolfram Burgard",
    "Dieter Fox",
    "Armin B. Cremers"
  ],
  "publication_date": "",
  "publication_venue": "The International Journal of Robotics Research",
  "abstract": "One of the goals in the field of mobile robotics is the development of mobile platforms which operate in populated environments. For many tasks it is therefore highly desirable that a robot can track the positions of the humans in its surrounding. In this paper we introduce sample-based joint probabilistic data association filters as a new algorithm to track multiple moving objects. Our method applies Bayesian filtering to adapt the tracking process to the number of objects in the perceptual range of the robot. The approach has been implemented and tested on a real robot using laser-range data. We present experiments illustrating that our algorithm is able to robustly keep track of multiple people. The experiments furthermore show that the approach outperforms other techniques developed so far.",
  "keywords": [
    "multi-target tracking",
    "data association",
    "particle filters",
    "people tracking",
    "mobile robot perception"
  ],
  "paper_institutions": [
    "University of Bonn, Computer Science Department, Germany",
    "University of Freiburg, Department of Computer Science, Germany",
    "University of Washington, Department of Computer Science & Engineering, Seattle, WA, USA"
  ],
  "output_language": "zh",
  "provider": "doubao",
  "model": "ark-code-latest",
  "prompt_template": "simple",
  "extracted_at": "2026-08-16T19:47:19.162971Z",
  "template_tag": "simple",
  "summary": "### 摘要\n本文针对移动机器人对多人进行跟踪的问题，提出了一种基于样本的联合概率数据关联滤波器（SJPDAF）。该方法将粒子滤波器的多模态密度表示能力与JPDAF的数据关联效率相结合，通过贝叶斯滤波动态自适应跟踪目标数量。系统利用激光测距数据，在真实机器人上实现了对多人的鲁棒跟踪。实验表明，该算法在人数变化、遮挡和杂波环境中均能准确跟踪，且在预测精度、关联正确率和抗杂波能力上优于标准JPDAF和其他现有方法。\n\n### 关键要点\n- 引入SJPDAF算法，利用粒子滤波表示每个目标状态，并通过JPDAF解决观测与目标的关联问题。\n- 通过维护目标数量的概率分布，采用递归贝叶斯更新来处理人数动态变化。\n- 基于激光扫描局部最小值提取腿部特征，并利用相邻扫描差异和占用栅格计算特征似然。\n- 通过遮挡图显式处理目标被遮挡的情况，避免样本集被错误删除。\n- 在真实机器人（Rhino）上进行实验，可实时处理3Hz的激光扫描数据。\n- 实验中机器人移动速度高达40cm/s，能跟踪多达6人，并处理动态进入和离开场景。\n- 与标准JPDAF相比，SJPDAF在预测中能表示双峰分布，避免了穿过障碍物的错误预测。\n- 在非线性运动模型中，SJPDAF避免了EKF的线性化误差，关联错误率显著降低。\n- 在高密度杂波环境中，SJPDAF的RMS误差远小于JPDAF，保持跟踪稳定性。\n- 当目标被遮挡时，单状态粒子滤波会丢失目标，而SJPDAF的独立样本集能保持跟踪。\n- 定量评估中，SJPDAF正确识别所有16条轨迹，平均位置误差16±5cm，检测延迟约1秒。\n- 遮挡处理实验显示，无遮挡处理时失败率17.5%，有遮挡处理时仅2.5%。"
}