Metraipolis › archive › dc_dd403fca83b1
Evaluating Physical Consistency and Plausibility in Generative Scenario Models for Autonomous Driving — arXiv:2610.01581
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.01581
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T12:05:42.871Z
- SHA-256 of the archived text
f0b366a037dfd846fa3eaf8aa4b5dbfc4620cfbea1154eef063d27d716398f74- Stable URL of this page
https://metraipolis.com/documento/dc_dd403fca83b1
How to cite
Evaluating Physical Consistency and Plausibility in Generative Scenario Models for Autonomous Driving — arXiv:2610.01581 — arXiv (en), https://arxiv.org/abs/2610.01581 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dd403fca83b1 (Metraipolis archive dc_dd403fca83b1, sha256 f0b366a037dfd846fa3eaf8aa4b5dbfc4620cfbea1154eef063d27d716398f74)
BibTeX
@misc{dcdd403fca83b1,
title = {Evaluating Physical Consistency and Plausibility in Generative Scenario Models for Autonomous Driving — arXiv:2610.01581},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.01581},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dd403fca83b1; sha256 f0b366a037dfd846fa3eaf8aa4b5dbfc4620cfbea1154eef063d27d716398f74},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.01581. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Generative AI models are increasingly used for scenario generation in autonomous driving. While they can generate realistic-looking scenarios, they often provide limited transparency into learned representations and consistency with real-world vehicle dynamics. This lack of formal assurance limits their use in safety-critical validation and certification workflows. To address this aspect, we introduce a layered evaluation protocol that complements existing methods by assessing models across five layers. The first four layers inspect internal representations and network layers through kinematic alignment, statistical baseline comparison, latent controllability, and activation analysis. The fifth layer evaluates model outputs against vehicle dynamics constraints such as lateral jerk thresholds. We demonstrate the protocol on a Variational Autoencoder (VAE)-based scenario generator. Although standard output-level metrics and visualizations suggest that the generated scenarios are realistic, our protocol provides deeper insight into the extent to which the model's latent space aligns with kinematic features and whether visually plausible trajectories satisfy vehicle-dynamics constraints. We further apply the protocol to additional generative models, demonstrating its applicability beyond the VAE architecture. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.01581 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-06T10:21:27.964Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).