PhysicalAgent：基于基础世界模型的通用认知机器人研究
PhysicalAgent: Towards General Cognitive Robotics with Foundation World Models
AI总结：
提出PhysicalAgent框架，结合迭代推理、扩散视频生成与闭环执行，通过视频演示与失败重规划实现机器人操作的稳健错误恢复，跨平台成功率达80%。
摘要：
我们提出了 PhysicalAgent，这是一个用于机器人操作的智能体框架，集成了迭代推理、基于扩散的视频生成和闭环执行。给定文本指令，我们的方法会生成候选轨迹的短视频演示，在机器人上执行它们，并根据失败情况迭代地重新规划。这种方法能够从执行错误中稳健地恢复。我们在多种感知模态（第一人称视角、第三人称视角和模拟）和机器人本体（双臂 UR3、Unitree G1 人形机器人、模拟 GR1）上评估了 PhysicalAgent，并与最先进的特定任务基线进行了比较。实验表明，我们的方法始终优于先前的方法，在人类熟悉的任务上成功率达到 83%。物理试验显示，首次尝试成功率有限（20-30%），但迭代纠正将跨平台的整体成功率提高到了 80%。这些结果突显了基于视频的生成推理在通用机器人操作中的潜力，并强调了迭代执行对于从初始失败中恢复的重要性。我们的框架为可扩展、适应性强且稳健的机器人控制铺平了道路。
英文摘要：
We introduce PhysicalAgent, an agentic framework for robotic manipulation that integrates iterative reasoning, diffusion-based video generation, and closed-loop execution. Given a textual instruction, our method generates short video demonstrations of candidate trajectories, executes them on the robot, and iteratively re-plans in response to failures. This approach enables robust recovery from execution errors. We evaluate PhysicalAgent across multiple perceptual modalities (egocentric, third-person, and simulated) and robotic embodiments (bimanual UR3, Unitree G1 humanoid, simulated GR1), comparing against state-of-the-art task-specific baselines. Experiments demonstrate that our method consistently outperforms prior approaches, achieving up to 83% success on human-familiar tasks. Physical trials reveal that first-attempt success is limited (20-30%), yet iterative correction increases overall success to 80% across platforms. These results highlight the potential of video-based generative reasoning for general-purpose robotic manipulation and underscore the importance of iterative execution for recovering from initial failures. Our framework paves the way for scalable, adaptable, and robust robot control.
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。
