Collecting Human Motion Data in Large and Occlusion-Prone Environments using Ultra-Wideband Localization

TL;DR

This work introduces a multi-modal data collection framework that leverages Ultra-Wideband localization to capture human motion in large, occlusion-prone spaces, using a museum-like environment to mimic real-world settings. By extending the THÖR protocol to include UWB tracking and integrating eye-tracking, LiDAR, radar, and motion capture, the authors assemble a comprehensive dataset with over 130 minutes of synchronized observations and ground-truth trajectories. The dataset enables direct evaluation of UWB accuracy against mocap and demonstrates the potential for radar-LiDAR fusion to map dynamic environments, even under occlusions. The work advances scalable, in-the-wild motion data collection and provides foundation for deploying UWB-based sensing in venues such as museums, warehouses, and airports.

Abstract

With robots increasingly integrating into human environments, understanding and predicting human motion is essential for safe and efficient interactions. Modern human motion and activity prediction approaches require high quality and quantity of data for training and evaluation, usually collected from motion capture systems, onboard or stationary sensors. Setting up these systems is challenging due to the intricate setup of hardware components, extensive calibration procedures, occlusions, and substantial costs. These constraints make deploying such systems in new and large environments difficult and limit their usability for in-the-wild measurements. In this paper we investigate the possibility to apply the novel Ultra-Wideband (UWB) localization technology as a scalable alternative for human motion capture in crowded and occlusion-prone environments. We include additional sensing modalities such as eye-tracking, onboard robot LiDAR and radar sensors, and record motion capture data as ground truth for evaluation and comparison. The environment imitates a museum setup, with up to four active participants navigating toward random goals in a natural way, and offers more than 130 minutes of multi-modal data. Our investigation provides a step toward scalable and accurate motion data collection beyond vision-based systems, laying a foundation for evaluating sensing modalities like UWB in larger and complex environments like warehouses, airports, or convention centers.

Figures (7)

• Figure 1: Overview of the datasets modalities and recording environment: The UWB system (1), eye-tracking glasses (2) and the robot with radar (3) and LiDAR (4) sensors.
• Figure 2: Top-down view of the indoor museum room setup. Layout II also shows the additional "Special Exhibit" area.
• Figure 3: Two UWB receiver devices used in the runs. Devices are equipped with IR-Markers for 6DoF MotionCapture
• Figure 4: The two obstacle layouts used in the recordings. They feature a few large obstacles that need to be navigated around (a) and many small obstacles that can be stepped over (b). Retractable banners introduce occlusions in the environment, adding visual search behavior to the scenario.
• Figure 5: Trajectories of four participants, recorded with the motion capture system. Results of a 3-minute run, colors corresponding to unique participants.