Summary
Sometimes robots can lose sight of the person they are following when something or someone blocks their view. This paper explains how the researchers combined camera and motion sensor data with information about the environment’s layout to help the robot guess where the person might be when out of sight. They used maps showing paths and intersections to predict likely routes a person would take. This way, the robot can keep moving safely and find the person again quickly once they come back into view. Tests showed the system works well in crowded places, finding the person again over 70% of the time even after 7 seconds of being hidden.
Autonomous mobile robotsPerson-followingOcclusionDeep learning-based visual trackingLiDAR point clusteringState estimationWheel odometryIMU sensorsTopological mapGraph-based search
Authors
Sagar Gupta, Don Gideon, Seng W. Loke, Kevin Lee, Bijo Sebastian
Abstract
Autonomous mobile robots performing person-following tasks often suffer from temporary occlusions and sensor track loss in dynamic environments. This research presents an end-to-end autonomous navigation stack that addresses target occlusion through map-informed spatial reasoning. The proposed system features a multi-modal perception pipeline, fusing deep learning-based visual tracking with 2-dimensional LiDAR point clustering to maintain high-fidelity tracking of a tagged person. A continuous state estimator integrates this perception data with wheel odometry and IMU sensors for stable localization. When the active track is lost due to occlusion, the system activates a map-based recovery framework. Leveraging a predefined topological map, the system executes a graph-based search to propagate the target's last known trajectory along structurally defined walking lanes, adhering to left-hand regional conventions. By generating a discrete set of feasible future trajectories, the robot reasons about potential structural trajectory changes, such as continuing a heading or turning at an intersection. This map-informed prediction is fed directly to the local obstacle avoidance planner, enabling the robot to continue following its target safely and predictably until the person is visually reacquired. Real-world evaluations in dense multi-person environments demonstrate the system's robustness, achieving a 71.4\% target reacquisition success rate during major occlusion events lasting up to 7 seconds.