AI Daybook
AI Daybook
오늘의 지면
논문Physical AI

PCBRouteAgent, PCB 배선 자동화 에이전트가 실제 보드에서 40~100% 성공률 달성

PCBRouteAgent: How Close Are LLM Agents to Becoming PCB Routing Engineers?

매체
Hugging Face
게시
중요도
3/5
화제 지수
11

한국어 요약

PCB 배선 자동화를 위해 글로벌 라우팅 계획과 상세 라우팅 실행을 조율하는 멀티모달 에이전트를 제시했다. 실제 PCB 보드 테스트 세트에서 40~100% 완전 성공률을 기록했고, 평가된 베이스라인 대비 13.9~36 포인트 높았다.

원문 발췌 · English

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.

#agent#multimodal#automation#pcb
원문 기사 읽기
전체 보기 →