推进仿人机器人运动：通过去噪世界模型学习掌握复杂地形
Advancing Humanoid Locomotion: Mastering Challenging Terrains with Denoising World Model Learning
AI总结：
本研究提出端到端强化学习框架DWL，通过去噪世界模型学习解决仿人机器人在复杂地形上的运动控制问题，实现全球首个掌握雪地、楼梯等挑战性地形的仿人机器人，并具备零样本仿真到现实迁移能力。
摘要：
仿人机器人凭借其类人骨骼结构，特别适合在以人为中心的环境中执行任务。然而，这种结构也给运动控制器设计带来了额外挑战，尤其是在复杂的现实环境中。因此，现有的仿人机器人仅限于相对简单的地形，无论是采用基于模型的控制还是无模型强化学习。在本研究中，我们提出了去噪世界模型学习（Denoising World Model Learning, DWL），这是一个用于仿人机器人运动控制的端到端强化学习框架。该框架展示了世界上首个能够掌握现实世界中复杂地形的仿人机器人，包括野外雪地和倾斜地面、上下楼梯以及极度不平坦的地形。所有场景均运行相同的已学习神经网络，并实现零样本仿真到现实迁移，这表明所提方法具有卓越的鲁棒性和泛化能力。
英文摘要：
Humanoid robots, with their human-like skeletal structure, are especially suited for tasks in human-centric environments. However, this structure is accompanied by additional challenges in locomotion controller design, especially in complex real-world environments. As a result, existing humanoid robots are limited to relatively simple terrains, either with model-based control or model-free reinforcement learning. In this work, we introduce Denoising World Model Learning (DWL), an end-to-end reinforcement learning framework for humanoid locomotion control, which demonstrates the world's first humanoid robot to master real-world challenging terrains such as snowy and inclined land in the wild, up and down stairs, and extremely uneven terrains. All scenarios run the same learned neural network with zero-shot sim-to-real transfer, indicating the superior robustness and generalization capability of the proposed method.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
