A Large-Scale Dataset for Humanoid Robotics Enabling a Novel Data-Driven Fall Prediction
In this paper, we present a comprehensive dataset comprising 37.9 hours of sensor data collected from humanoid robots, including 18.3 hours of walking and 2,519 recorded falls. This extensive dataset is a valuable resource for various robotics and machine learning applications. Leveraging this data, we propose {RePro}-{TCN}, a Temporal Convolutional Network ({TCN}) enhanced with two novel extensions: Relaxed Loss Formulation and Progressive Forecasting. Predicting falls is a critical capability in humanoid robotics for implementing countermeasures such as lunging or stopping the walk. Thanks to the new dataset, we train {RePro}-{TCN} and demonstrate its superiority over previous approaches under real-world conditions that were previously unattainable.
- Published in:
2025 IEEE International Conference on Robotics and Automation (ICRA) - Type:
Inproceedings - Year:
2025 - Source:
https://ieeexplore.ieee.org/abstract/document/11128646
Citation information
: A Large-Scale Dataset for Humanoid Robotics Enabling a Novel Data-Driven Fall Prediction, 2025 IEEE International Conference on Robotics and Automation (ICRA), 2025, 13906--13912, May, IEEE, https://ieeexplore.ieee.org/abstract/document/11128646, Urbann.etal.2025a,
@Inproceedings{Urbann.etal.2025a,
author={Urbann, Oliver; Eßer, Julian; Kleingarn, Diana; Moos, Arne; Brämer, Dominik; Brömmel, Piet; Bach, Nicolas; Jestel, Christian; Larisch, Aaron; Kirchheim, Alice},
title={A Large-Scale Dataset for Humanoid Robotics Enabling a Novel Data-Driven Fall Prediction},
booktitle={2025 IEEE International Conference on Robotics and Automation (ICRA)},
pages={13906--13912},
month={May},
publisher={IEEE},
url={https://ieeexplore.ieee.org/abstract/document/11128646},
year={2025},
abstract={In this paper, we present a comprehensive dataset comprising 37.9 hours of sensor data collected from humanoid robots, including 18.3 hours of walking and 2,519 recorded falls. This extensive dataset is a valuable resource for various robotics and machine learning applications. Leveraging this data, we propose {RePro}-{TCN}, a Temporal Convolutional Network ({TCN}) enhanced with two novel...}}