arXiv cs.AI· Elias Hofele, Thomas Trapp, David Timmermann, Tristan Schnell, Georg Heppner, Arne Roennau·· 4 小时前AI 评分35
TrySee：面向专用领域闭环机器人任务规划的模块化感知增强 VLM 具备接地框架
TrySee: Modular Perception-Enhanced VLM Grounding for Closed-Loop Robot Task Planning in Specialized Domains
AI 导读
TrySee 是一个模块化闭环机器人任务规划管线，将通用 VLM 推理与领域专用感知及成熟机器人组件结合。系统无需 VLM 微调或端到端演示，通过可替换感知模块和运动规划进行适配，并借助动态预处理选择机制实现扩展。该方法在 5 个差异显著的领域上评估，包括真实世界机器人拆解任务，brownfield 环境现场测试证明了其在领域训练数据有限时的实用性。
正文 · 原文
Abstract:Vision Language Models (VLMs) have shown impressive capabilities in high-level reasoning and task planning for robots. However, specialized domains remain a challenge for VLMs' image interpretation and grounding abilities. We introduce TrySee, a modular closed-loop robot task planning pipeline integrating general-purpose VLM reasoning with domain-specific perception and established robotics components. The system consists of an inherently reactive cyclic planning pipeline, augmented with domain-specific modular vision preprocessing, to provide online grounding. Finally, a dynamic preprocessing selection mechanism contributes to the scalability of the pipeline. The system requires neither VLM fine-tuning nor end-to-end demonstrations and adapts through replaceable perception modules and motion planning. The approach is evaluated on planning tasks across 5 widely different domains, including a real-world robot disassembly task. The field tests in a brownfield environment demonstrate the approach's practicality in integrating VLM planning into specialized robotic systems with limited domain-specific training data.
| Comments: | To be published in the proceedings of the 2026 IEEE International Conference on Robotics and Biomimetics (ROBIO) |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.35687 [cs.RO] |
| (or arXiv:2610.35687v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.35687 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Elias Hofele [view email]
[v1]
Mon, 26 Oct 2026 16:20:28 UTC (12,504 KB)
来源：arXiv cs.AI · arxiv.org