A Large-Scale Dataset for Simulated Car-Pedestrian Collisions to Advance Safety Research
| dc.contributor.author | Medina, Edgar | |
| dc.contributor.author | Loh, Leyong | |
| dc.contributor.editor | Skala, Václav | |
| dc.date.accessioned | 2025-07-30T10:04:31Z | |
| dc.date.available | 2025-07-30T10:04:31Z | |
| dc.date.issued | 2025 | |
| dc.description.abstract-translated | Car-pedestrian collisions are a daily occurrence worldwide, yet there is a notable absence of public datasets in this domain. Research in this area is crucial, as it directly impacts pedestrian safety and serves as a basis for validating autonomous driving systems. Although finite element simulations are used, they are computationally intensive and yield insufficient data for deep learning applications. In this work, we present the PeDesCar dataset for safe autonomous driving, which spans around 15 days of simulated time and encompasses over 1 million collision events, each constrained within a temporal window of up to 2 seconds per event. The dataset is generated using MuJoCo as a physics simulator, proving its effectiveness in sim2real robotics research. We use PeDesCar to train and assess state-of-the-art models in human motion prediction and validate the realism of the simulation against realistic high-fidelity finite element simulations. Our results validate that PeDesCar is sufficient for preliminary car-pedestrian collision research. The visualization code and dataset are accessible on the project website https: //github.com/QualityMinds/PeDesCar. | en |
| dc.description.sponsorship | The research leading to these results is funded by the German Federal Ministry for Economic Affairs and Climate Action within the project “ATTENTION – Artificial Intelligence for realtime injury prediction”. The authors appreciate the consortium’s successful collaboration and extend special thanks to Niels Heller for his contributions to data management. | |
| dc.format | 14 s. | cs |
| dc.format.mimetype | application/pdf | |
| dc.identifier.doi | http://www.doi.org/10.24132/CSRN.2025-21 | |
| dc.identifier.issn | 2464-4617 (Print) | |
| dc.identifier.issn | 2464-4625 (online) | |
| dc.identifier.uri | http://hdl.handle.net/11025/62227 | |
| dc.language.iso | en | en |
| dc.publisher | Vaclav Skala - UNION Agency | en |
| dc.rights | © Vaclav Skala - UNION Agency | en |
| dc.rights.access | openAccess | en |
| dc.subject | PeDesCar | cs |
| dc.subject | predikce lidského pohybu | cs |
| dc.subject | simulace | cs |
| dc.subject.translated | PeDesCar | en |
| dc.subject.translated | human motion prediction | en |
| dc.subject.translated | simulation | en |
| dc.title | A Large-Scale Dataset for Simulated Car-Pedestrian Collisions to Advance Safety Research | en |
| dc.type | konferenční příspěvek | cs |
| dc.type | conferenceObject | en |
| dc.type.status | Peer reviewed | en |
| dc.type.version | publishedVersion | en |
| local.files.count | 1 | * |
| local.files.size | 4128521 | * |
| local.has.files | yes | * |