AI Daybook
AI Daybook
오늘의 지면
논문Physical AI

HexaShell, 로봇 코딩 에이전트용 범용 하네스 프레임워크

HexaShell: A Universal Harness for Robot Coding Agents in Simulation and the Real World

매체
arXiv
게시
중요도
2/5
화제 지수
3

한국어 요약

VLM 기반 로봇 코딩 에이전트를 위한 범용 하네스인 HexaShell을 제시했다. 14개의 기본 도구(동작 4개, 인지 5개, 상태 5개)로 구성되며, 네 개의 시뮬레이션 벤치마크와 실제 로봇에서 동일한 코드로 작동한다. GPT-6-Astra와 작업별 입력 없이 RoboDojo에서 69.9%, RoboCasa365에서 59.3%, LIBERO-PRO에서 78.3%, RoboTwin 2.0에서 82.2%의 성공률을 달성했다.

원문 발췌 · English

Agents built on vision language models (VLMs) now control robots by writing code, but the strongest of them refine task-specific code over many rounds of trial and revision. Such code tends to overfit the benchmark on which it was tuned, can exploit the benchmark's shortcuts, and is costly to carry over to a new benchmark or robot. We present HexaShell, a thin, task-agnostic harness designed to test how far general tools can go instead. We design a simple and complete set of 14 basic tools, 4 for motion, 5 for perception and 5 for state, which the agent calls in one Python cell per turn. Perception pairs SAM 3, which segments an object from a few words or a click, with NumPy and OpenCV geometry that turns its outlines in several cameras, or a depth image, into its position, size and grasp angle. The API, agent loop and sandbox are shared across all tasks; a new simulator or robot needs only an adapter of 350 to 1,450 lines and a prompt describing its world, and the only task-specific input is an optional skill of about a page of guidance. The same code runs on four simulated benchmarks, built on three simulators with two kinds of arm, and on real robots. With GPT-6-Astra and no task-specific input, HexaShell reaches success rates of 69.9% on RoboDojo, 59.3% on RoboCasa365, 78.3% on LIBERO-PRO and 82.2% on RoboTwin 2.0. With task skills, it reaches 85.0% (score 88.8), 85.6%, 96.7% and 96.8%, respectively, above the best published result we are aware of on each benchmark. The RoboCasa365 result is listed first on its leaderboard. On real dual-arm robots, HexaShell succeeds in 76 of 90 episodes across nine tasks.

#vlm#coding-agent#robotics
원문 기사 읽기
전체 보기 →