外观
论文arXiv 2610.1242420 天前
RoboRSI：复杂真实环境中稳定、高效且可复用的机器人自我进化
原文链接：RoboRSI: Stable, efficient, and reusable robot self-evolution in complex real-world environments（arXiv 2610.12424 · 发布日期：2026-10-08）
RoboRSI: Stable, efficient, and reusable robot self-evolution in complex real-world environments
RoboRSI用Top-Down Skill Refinement组织执行经验，在移动操作机器人和多个仿真基准上验证机器人自我改进。
核心要点
- 面向通用机器人从执行经验中改进并复用能力的问题，关键挑战是按任务结构归因、验证和组织修复。
- TSR将任务分解为复合、原子和基础技能，定义职责与输入输出契约，并把修订限制在负责分支。
- Manager、Planner、Engineer和Reviewer协同规划、执行、诊断与技能发布，人类通过目标和纠正引导过程。
- RoboRSI在移动操作机器人上发展多物体家庭清理；仿真中在多个基准取得最高成功率并超过最强基线。
验证范围与局限：验证范围包括移动操作机器人上的104轮多物体家庭清理，以及LIBERO、LIBERO-PRO、LIBERO-Plus和RoboTwin仿真。
英文摘要 (Abstract)
A generalist robot should not only perform diverse tasks but also improve through experience, turning what it learns during execution into capabilities that later tasks can reuse. Robot agents that act through code can already repair programs from execution feedback, yet it remains a central challenge to organize this experience around the task structure that gives it meaning, so that each repair is attributed to the responsible capability, supported by execution evidence, and validated before it is reused. We introduce RoboRSI, a robot self-improvement system built on Top-Down Skill Refinement (TSR). TSR decomposes tasks into compound, atomic, and base skills with scoped responsibilities and explicit input--output contracts, attributes each execution outcome to the responsible branch, and confines revision to that branch. Building upon this structure, a Manager, Planner, Engineer, and Reviewer coordinate planning, execution, diagnosis, and the validated release of new skills, while people steer the process through objectives and corrections; stable skill sequences are further consolidated into reusable compound skills. On a mobile manipulator, RoboRSI develops multi-object household cleanup over 104 rounds. In simulation, it achieves the highest success rate on LIBERO, LIBERO-PRO, LIBERO-Plus, and RoboTwin, exceeding the strongest baseline by 2.7 to 11.0 percentage points.
作者：Zimo Wen, Yijin Chen, Yuxuan Cao, Wendi Chen, Yanwen Zou, Wenye Yu, Fuhang Kuang, Han Xue, Jun Lv, Chuan Wen
来源
本页事实对应的公开页面，以原文为准。