階層型ワールドモデルによる視覚ベース全身ヒューマノイド制御
Hierarchical World Models as Visual Whole-Body Humanoid Controllers
視覚観測から56自由度ヒューマノイドの全身制御を強化学習で実現するため、高レベルエージェントが視覚に基づきコマンドを生成し低レベルエージェントが実行する階層型ワールドモデルを提案した。
著者: Nicklas Hansen, Jyothir S, Vlad Sobal, Yann LeCun, Xiaolong Wang, Hao Su
分類: cs.LG, cs.CV, cs.RO
原文アブストラクト
Whole-body control for humanoids is challenging due to the high-dimensional nature of the problem, coupled with the inherent instability of a bipedal morphology. Learning from visual observations further exacerbates this difficulty. In this work, we explore highly data-driven approaches to visual whole-body humanoid control based on reinforcement learning, without any simplifying assumptions, reward design, or skill primitives. Specifically, we propose a hierarchical world model in which a high-level agent generates commands based on visual observations for a low-level agent to execute, both of which are trained with rewards. Our approach produces highly performant control policies in 8 tasks with a simulated 56-DoF humanoid, while synthesizing motions that are broadly preferred by humans.
