论文概要
研究领域: ML 作者: Wancong Zhang, Basile Terver, Artem Zholus 等 发布时间: 2026-04-03 arXiv: 2604.03208
中文摘要
基于学习世界模型的模型预测控制已成为具身控制的有前景范式。然而，学习的世界模型往往难以处理长程控制，原因是预测误差的累积和指数增长的搜索空间。本文通过在多时间尺度上学习潜在世界模型并在这些尺度上进行分层规划来解决这些挑战。我们证明这种分层方法能够在真实世界非贪心机器人任务上实现零样本控制，仅使用最终目标规范在拾取放置任务上达到70%成功率，而单层世界模型为0%。
原文摘要
Model predictive control (MPC) with learned world models has emerged as a promising paradigm for embodied control, particularly for its ability to generalize zero-shot when deployed in new environments. However, learned world models often struggle with long-horizon control due to the accumulation of prediction errors and the exponentially growing search space. In this work, we address these challenges by learning latent world models at multiple temporal scales and performing hierarchical planning across these scales, enabling long-horizon reasoning while substantially reducing inference-time planning complexity. Our approach serves as a modular planning abstraction that applies across diverse latent world-model architectures and domains. We demonstrate that this hierarchical approach enabl...
--- *自动采集于 2026-04-06*
#论文 #arXiv #ML #小凯