arXiv cs.AI· Sad{\i}k Bera Y\"uksel, Derya Aksaray·· 11 小时前AI 评分31
面向机器人基础模型的规范感知动作分布优化框架
Specification-Aware Distribution Shaping for Robotics Foundation Models
AI 导读
研究者提出一种规范感知的动作分布优化框架，可在不修改预训练机器人基础模型参数的前提下，于执行阶段强制满足一类 Signal Temporal Logic（STL）约束。该方法在每个决策步通过前向动力学传播推理剩余时域，计算出满足硬性 STL 可行性约束的最小修改动作分布。研究在仿真中基于当前先进的机器人基础模型、跨多个环境和复杂规范完成了验证。
正文 · 原文
Abstract:Robotics foundation models have demonstrated strong capabilities in executing natural language instructions across diverse tasks and environments. However, they remain largely data-driven and lack formal guarantees on constraint satisfaction during deployment. In practice, robots often need to comply with operational constraints involving rich spatio-temporal requirements such as time-bounded goal visits, sequential objectives, and persistent safety conditions. In this work, we propose a specification-aware action distribution optimization framework that enforces a broad class of Signal Temporal Logic (STL) constraints during execution of a pretrained robotics foundation model without modifying its parameters. At each decision step, the method computes a minimally modified action distribution that satisfies a hard STL feasibility constraint by reasoning over the remaining horizon using forward dynamics propagation. We validate the proposed framework in simulation using state-of-the-art robotics foundation models across multiple environments and complex specifications.
| Comments: | 8 pages, 4 figures |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2603.17969 [cs.RO] |
| (or arXiv:2603.17969v2 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2603.17969 arXiv-issued DOI via DataCite |
Submission history
From: Sadık Bera Yüksel [view email]
[v1]
Wed, 18 Mar 2026 17:36:46 UTC (4,865 KB)
[v2]
Fri, 30 Oct 2026 15:28:37 UTC (3,294 KB)
来源：arXiv cs.AI · arxiv.org