Metraipolis › archive › dc_4f250ede45cd
Learning to Explain While Planning: Rule-Aligned Diffusion Planning for Autonomous Driving — arXiv:2609.39995
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.39995
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T13:46:49.737Z
- SHA-256 of the archived text
237ad48eee6cbd0c2d21efaf20d499eba87d160ab5e5958b32103c2c42825c1d- Stable URL of this page
https://metraipolis.com/documento/dc_4f250ede45cd
How to cite
Learning to Explain While Planning: Rule-Aligned Diffusion Planning for Autonomous Driving — arXiv:2609.39995 — arXiv (en), https://arxiv.org/abs/2609.39995 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4f250ede45cd (Metraipolis archive dc_4f250ede45cd, sha256 237ad48eee6cbd0c2d21efaf20d499eba87d160ab5e5958b32103c2c42825c1d)
BibTeX
@misc{dc4f250ede45cd,
title = {Learning to Explain While Planning: Rule-Aligned Diffusion Planning for Autonomous Driving — arXiv:2609.39995},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.39995},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4f250ede45cd; sha256 237ad48eee6cbd0c2d21efaf20d499eba87d160ab5e5958b32103c2c42825c1d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39995. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Diffusion planners exhibit strong capabilities in generating multimodal trajectories. However, existing methods primarily rely on expert demonstrations to fit trajectory distributions, learning statistical correlations among scenes, behaviors, and trajectories without explicitly modeling driving rules. In long-tail scenarios where expert data are scarce, the lack of behaviors to imitate may lead to trajectories that violate safety or compliance requirements. Moreover, their generation process lacks rule-level explanations, making it difficult to determine which rules drive trajectory adjustments, when they take effect, and how strongly they act, thereby limiting failure diagnosis, safety validation, and targeted improvement. To address these limitations, we propose the Rule-Aligned Diffusion Planner (RADP), which incorporates differentiable driving rules into the diffusion objective during training, turning rule knowledge into intrinsic behavioral principles beyond finite demonstrations. We further introduce Rule-Pressure Attribution (RPA), which constructs supervision signals from gradients of rule losses with respect to predicted trajectories and employs a lightweight attribution head to estimate the optimization pressure exerted by each rule online. To assess the closed-loop behavioral relevance of these attributions, we propose a temporal risk-alignment protocol that evaluates whether current rule pressures reflect corresponding risks during subsequent closed-loop execution. Experiments on nuPlan show that RADP improves closed-loop planning in challenging safety-critical scenarios, while RPA exhibits consistent temporal alignment with subsequent rule-specific risks, validating both intrinsic rule learning and rule-level interpretability. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.39995 — 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 2026-10-07T13:18:10.495Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).