基于UWB的密集人流区域室内定位跟随算法研究

Research on Indoor Positioning and Following Algorithm for Dense Crowd Areas Based on UWB

摘要: 在当下的零售环境中，传统的超宽带跟随系统存在两大核心难题：一是必须让人佩戴标签才能定位，不够灵活；二是它对那些没戴标签的顾客“视而不见”，在人头攒动的商场里，这就等于有了感知盲区，机器人很容易跟丢目标，或者撞上突然出现的人。为了让UWB既能“看见”戴了标签的人，也能感知到周围没戴标签的人群，文章提出了一种“有源精准定位”与“无源群体感知”相协同的混合UWB感知新范式及其关键算法。研究创新性地设计了一套复用同一套UWB硬件的新框架，该框架具备双重功能：其一，通过优化TDOA (到达时间差)算法，结合环境感知来消除障碍物的干扰，为佩戴标签的目标提供厘米级的稳定跟踪；其二，通过解析无线信号在空间传播中留下的痕迹(信道状态信息，CSI)，去反推周围人群的密度和整体的流动趋势。这就像同时拥有了高清摄像头和热力图，既能看到清晰的个体，也能感知到周围环境的温度。基于这一思路，通过系统性的理论分析、模型构建与方案设计，阐述了问题定义、架构创新与算法研究的逻辑闭环，为解决密集动态场景下智能服务机器人的精准跟随与安全交互问题提供了全新的技术路径，具有重要的理论意义与应用前景。

Abstract: In the current retail environment, the traditional ultra-wideband tracking system faces two core challenges: first, it requires people to wear tags for positioning, which is not flexible enough; second, it ignores customers who do not wear tags, resulting in a perception blind spot in crowded shopping malls. This makes it easy for robots to lose track of targets or collide with unexpected individuals. To enable UWB to “see” both those wearing tags and the surrounding untagged crowd, this paper proposes a hybrid UWB perception paradigm that combines “active precise positioning” and “passive group sensing” and its key algorithms. This study innovatively designs a new framework that reuses the same UWB hardware, which has two functions: first, by optimizing the TDOA (Time Difference of Arrival) algorithm and combining environmental perception to eliminate the interference of obstacles, it provides centimeter-level stable tracking for the tagged targets; second, by analyzing the traces left by wireless signals during their propagation in space (Channel State Information, CSI), it deduces the density and overall flow trend of the surrounding crowd. This is like having both a high-definition camera and a heat map, allowing one to see clear individuals while also sensing the temperature of the surrounding environment. Based on this idea, this study systematically conducts theoretical analysis, model construction, and scheme design to elaborate on the logical loop of problem definition, architecture innovation, and algorithm research, providing a new technical path for solving the precise following and safe interaction problems of intelligent service robots in dense dynamic scenarios, and has significant theoretical significance and application prospects.