Metraipolis › archive › dc_8ea061dc2621

TRACE: Target-State Risk-Aligned Conditional Execution for Embodied Tracking — arXiv:2610.41275

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.41275
License (as archived)
CC0-1.0
Archived
2026-11-02T05:05:37.870Z
SHA-256 of the archived text
019dad4f1bf0679d17d856d8d151ad471603b531739e60ff2c9e9459ea9525f3
Stable URL of this page
https://metraipolis.com/documento/dc_8ea061dc2621

How to cite

TRACE: Target-State Risk-Aligned Conditional Execution for Embodied Tracking — arXiv:2610.41275 — arXiv (en), https://arxiv.org/abs/2610.41275 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8ea061dc2621 (Metraipolis archive dc_8ea061dc2621, sha256 019dad4f1bf0679d17d856d8d151ad471603b531739e60ff2c9e9459ea9525f3)
BibTeX
@misc{dc8ea061dc2621,
  title = {TRACE: Target-State Risk-Aligned Conditional Execution for Embodied Tracking — arXiv:2610.41275},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41275},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8ea061dc2621; sha256 019dad4f1bf0679d17d856d8d151ad471603b531739e60ff2c9e9459ea9525f3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41275. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Embodied visual tracking requires a robot to follow a language-specified person under occlusion, look-alike distractors, and ambiguous instructions. We show that improving tracking persistence can simultaneously increase safety violations. Using a compact 0.6B vision-language-action (VLA) tracker with a frozen backbone, we update only the planner head with proximal policy optimization (PPO). PPO improves success and following rates across all three EVT-Bench settings, but increases the collision rate on distracted tracking because episodes that previously ended in target loss can instead end in contact. We introduce TRACE, an execution-time target-state risk gate that preserves the learned policy and intervenes selectively. A lightweight risk head processes 84 cues, including candidate boxes, identity margins, a learned range-and-bearing probe, and the policy's commands. When the calibrated risk exceeds a threshold, TRACE rescales the translational component of the next waypoint while leaving yaw unchanged. The executed action is identical to PPO's on more than 98% of steps. In paired evaluations on identical episodes, TRACE reduces the distracted-tracking collision rate by 3.96 percentage points, while further improving success and following rates. Matched slowdown controls without target-state timing do not reproduce this improvement. Concurrent evaluation reproduces a collision reduction of 3.77 points, compared with 0.29 points for identical-arm null pairs, indicating that the effect exceeds evaluation variability. These results show that target-state-aware execution can recover the safety cost of policy post-training with minimal intervention.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.RO, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2610.41275 — 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 2026-11-02T04:49:28.194Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).