[discussion] Physical AI sim-real track A2-S → A4-S — review & gaps · Physical AI sim-real 트랙 검토

[github-actions]

[discussion] Physical AI sim-real track A2-S → A4-S — review & gaps · Physical AI sim-real 트랙 검토

Labels: rfc, help wanted, docs, area:runtime, area:bridge-python, priority:medium

Related: Closes scope gap after #28 · RFC-0001 · PR feat/physical-ai-samples-a3-a4

---

Goal · 목표

Community review of the sim-real Physical AI reference path (not only motion-alarm PoC):

Community가 sim-real Physical AI 레퍼런스 경로를 검토합니다 (motion-alarm PoC만이 아님):

Tier	Path	Inject
A2-S	examples/spiderling-sim/	ImuSample
A3 / A3-S / A3-H	examples/spider-robot/ · spider-robot-sim/	VisionFrame
A4 / A4-S / A4-H	examples/pet-robot/ · pet-robot-sim/	OwnerPing

Docs: bridge-python/ROBOTICS.md, rfcs/RFC-0001-stream-membrane-physical-sla.md

---

Questions for Discussion · Discussion 질문

1. Is the linear teaching trace (declared streams unused on path) honest enough for learners?
   linear teaching trace(선언 stream 미사용)가 학습용으로 충분히 정직한가?
2. Are SCENARIO / A3-H / A4-H EN/KR docs clear for international contributors?
   SCENARIO / A3-H / A4-H EN/KR 문서가 국제 기여자에게 명확한가?
3. What should A5 Humanoid borrow from this pattern (streams, SLA entry cell)?
   A5 Humanoid가 이 패턴(stream, SLA entry)에서 무엇을 가져가야 하는가?

---

Known limits (documented, need consensus) · 알려진 한계 (합의 필요)

• Subprocess cell run per bridge pulse (no daemon) · pulse마다 subprocess
• No multi-stream parallel inject in one session · 단일 세션 multi-stream inject 없음
• PET TouchStream teaching-only · PET TouchStream teaching-only
• DSL has no binary - — samples use if/else branches · DSL - 연산 없음

---

Tasks · 작업 (optional PRs)

• Link learning path from root README.md to A4-S / A4-H after merge · README 학습 경로 링크
• Add cell:test sidecar note for pet-sim-organism.cell in examples/README.md · cell:test 문서
• Viewer traces for PET comfort/follow paths · Viewer trace 추가
• Respond in Discussion with agreed “won’t fix vs roadmap” for each known limit · Discussion에 한계 합의

---

Definition of done · 완료 기준

• Discussion thread has maintainer summary: accepted teaching simplifications vs planned issues
  Discussion에 maintainer 요약: 수용한 teaching simplification vs 계획된 이슈
• At least one external contributor comment or reaction
  외부 기여자 코멘트 또는 reaction 1건 이상
• Remaining work split into trackable issues (20, 21, …)
  잔여 작업이 추적 가능한 이슈로 분리됨

---

References · 참고

• examples/pet-robot/SCENARIO.md
• examples/pet-robot-sim/SCENARIO.md
• examples/spider-robot-sim/SCENARIO.md
• .github/discussions/003-physical-ai-samples-feedback.md