arXiv cs.AI· Yue Zhang, Jilei Sun, Liqiang Jing, Yijie Wei, Ruohan Yang·· 12 小时前AI 评分34
PCBRouteAgent：LLM 智能体距离成为 PCB 布线工程师还有多远？
PCBRouteAgent: How Close Are LLM Agents to Becoming PCB Routing Engineers?
AI 导读
多模态 PCB 布线智能体 PCBRouteAgent 通过全局布线规划器结合板图与结构化电气几何信息，再由详细布线执行器调用 KiCad 原生布线引擎完成局部恢复。在 PCBWorld-Bench 四个测试划分共 247 块合成与真实电路板上，其 clean-pass 率分别为 100%、94.9%、100% 和 40%；三个真实场景子集上比最佳基线分别高出 13.9、36 和 10 个百分点。
正文 · 原文
Abstract:Automating printed circuit board (PCB) routing requires coordinating board-level connectivity decisions with geometric execution under strict design rules. Although recent agents can interact with native routing tools, reliable routing on complex boards remains challenging. We present PCBRouteAgent, a multimodal PCB routing agent inspired by how human engineers coordinate global strategy with local execution. A Global Routing Planner combines board images and structured electrical and geometric information to specify connection topology, routing order, corridors, and layer transitions. A Detailed Routing Executor realizes these guides through KiCad's native routing engine and performs bounded local recovery. Unresolved failures are returned as structured feedback, enabling targeted guide revisions while preserving accepted routes. We evaluate PCBRouteAgent on 247 synthetic and real-world boards from four PCBWorld-Bench test splits. It achieves clean-pass rates of 100%, 94.9%, 100%, and 40% on D2, D3-A, D3-B, and D3-C, respectively. On the three real-world subsets, PCBRouteAgent outperforms the best evaluated baseline in clean-pass rate by 13.9, 36, and 10 percentage points, respectively.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.08175 [cs.AI] |
| (or arXiv:2611.08175v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.08175 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Liqiang Jing [view email]
[v1]
Thu, 5 Nov 2026 15:31:15 UTC (1,138 KB)
来源：arXiv cs.AI · arxiv.org