Metraipolis › archive › dc_6ce5ec89d2fa

ConDR: Conservative Reachability through Neural Hamilton-Jacobi Subsolutions — arXiv:2611.06802

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.06802
License (as archived)
CC0-1.0
Archived
2026-11-05T18:22:44.651Z
SHA-256 of the archived text
98176a578a1c9927c6602fe56c649ea76a95e37b849705a7c2da6d9167d845e6
Stable URL of this page
https://metraipolis.com/documento/dc_6ce5ec89d2fa

How to cite

ConDR: Conservative Reachability through Neural Hamilton-Jacobi Subsolutions — arXiv:2611.06802 — arXiv (en), https://arxiv.org/abs/2611.06802 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6ce5ec89d2fa (Metraipolis archive dc_6ce5ec89d2fa, sha256 98176a578a1c9927c6602fe56c649ea76a95e37b849705a7c2da6d9167d845e6)
BibTeX
@misc{dc6ce5ec89d2fa,
  title = {ConDR: Conservative Reachability through Neural Hamilton-Jacobi Subsolutions — arXiv:2611.06802},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.06802},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6ce5ec89d2fa; sha256 98176a578a1c9927c6602fe56c649ea76a95e37b849705a7c2da6d9167d845e6},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.06802. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Hamilton-Jacobi (HJ) Reachability identifies the unsafe states of a dynamical system from which it can be forced into failure under worst-case disturbance. However, classical grid-based solvers scale exponentially with state dimension. DeepReach improves scalability by replacing the grid with a neural value function trained on the HJ-Isaacs (HJI) variational inequality (VI), but penalizes both sides of the residual equally. This allows the learned value to lie above the true one and unsafe states to be misclassified as safe. We propose Conservative DeepReach (ConDR) which aims to learn a subsolution of the HJI VI such that the learned unsafe set contains the true one. ConDR enforces the obstacle branch and terminal condition by construction, pushes the HJI residual to a safety margin $\epsilon$ below zero with a one-sided loss, and prevents the degenerate solution of classifying everything as unsafe. We prove that the exact solution of the margin-shifted VI that ConDR trains toward is a subsolution of the original VI so its unsafe set contains the true one and its value lies within $\epsilon\tau$ of the true value function. We compare against DeepReach and MADR in the established systems of Air3D, Air6D, a 9D multi-vehicle system, and a 13D quadrotor. ConDR reports the lowest rate of unsafe state misclassified and the lowest closed-loop collision rates with at most a small increase in intervention. On Air3D and Air6D, ConDR tolerates model mismatch that scales with $\epsilon$. At matched false-negative rates, ConDR's unsafe sets are smaller than the baselines showing the safety is not obtained by trivial over-approximation.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO) ; Machine Learning (cs.LG); Systems and Control (eess.SY).
Pagina del lavoro: https://arxiv.org/abs/2611.06802 — 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 2026-11-05T18:15:49.042Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).