Leveraging Pedestrian Detection and Tracking in Robotics Navigation: A Survey With Practical Illustrations

Pedestrian Detection and Tracking (PDT) plays a pivotal role in enabling autonomous robots to navigate safely and efficiently in dynamic, human-populated environments. This paper presents a comprehensive survey of PDT methods, structured according to the sensing modalities employed: RGB cameras, LiDAR, thermal imaging, RGB-D sensors, and multi-modal fusion systems. For each category, we analyze representative techniques, synthesize their strengths and limitations, and discuss recent advancements including deep learning approaches and cross-modal fusion strategies. We highlight persistent challenges such as handling occlusions, achieving real-time performance, and ensuring robustness across diverse environments. In addition to this structured review, we provide two practical examples using the Ona autonomous robot platform (see <xref ref-type="fig" rid="fig1" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Figure 1</xref>) to illustrate how PDT techniques can enhance robotic capabilities in real-world scenarios. These examples focus on improving SLAM consistency and enabling proxemic-aware navigation strategies. Through this survey, we aim to clarify the current state of the art, identify emerging trends, and suggest future research directions for robust and socially-aware robotic navigation.