arXiv：cs.AI· Angen Ye, junjie xie, Yujia Song, Junkai Huang, Dapeng Zhang·· 18 小时前
SurgiCLAW：面向精密机器人显微操作的可解释原子技能智能体
SurgiCLAW: An Interpretable Atomic-Skill Agent for Precision Robotic Micromanipulation
AI 导读
SurgiCLAW 是一个分层技能框架，将 MLLM 限定于术中视觉理解、结构化解释和原子手术技能选择，物理执行交由可验证的确定性控制程序完成。在脑机接口植入任务中，该框架在固定测试集上实现 94.6% 技能选择准确率和 93.0% Macro-F1，50 次闭环真机植入试验任务成功率达 88.0%。
正文
Abstract:Recent multimodal large language models (MLLMs) and vision--language--action models have provided robots with new capabilities for long-horizon task understanding and planning. In surgical robotics, however, end-to-end prediction of low-level actions remains limited by insufficient precision, opaque decisions, and scarce data. This paper presents SurgiCLAW, a hierarchical skill-based framework for precision surgical manipulation. SurgiCLAW confines an MLLM to intraoperative visual understanding, structured explanation, and atomic surgical-skill selection, while delegating physical execution to verifiable deterministic control programs. For brain--machine-interface implantation, a representative high-precision surgical task, we construct a state-to-skill invocation dataset and expand it through visual and state perturbations. The model first learns structured skill invocation through supervised fine-tuning and is then optimized using three rewards for action correctness, operation prerequisites, and format validity, improving action selection, operational safety, and output parseability. We evaluate subtask success, offline model comparison, real-robot performance, and ablations. On a fixed test set, SurgiCLAW achieves 94.6% skill-selection accuracy and 93.0% macro-averaged F1 (Macro-F1). Across 50 closed-loop real-robot implantation trials, it achieves an 88.0% task success rate.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.15766 [cs.RO] |
| (or arXiv:2610.15766v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.15766 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Angen Ye [view email]
[v1]
Sun, 11 Oct 2026 16:45:21 UTC (3,165 KB)
来源：arXiv：cs.AI · arxiv.org