近日，北京大学李廉林及其研究团队取得一项新进展。他们提出了电磁超材料智能体。相关研究成果已于2025年1月1日在国际知名学术期刊《光：科学与应用》上发表。

该研究团队提出并实验性地构建了一个范式转变，即打造一种具备推理和认知能力的超材料智能体（命名为“超智能体”），使其能够自主规划并成功执行多种长期任务，包括电磁（EM）场操控以及与机器人和人类的交互。超智能体利用最新发布的基础模型，以高级自然语言进行推理，并对来自不断演变的复杂环境的各种提示作出响应。

具体而言，超智能体的“大脑”通过多智能体讨论机制，以自然语言进行高级任务规划，其中各个智能体分别是感知、规划、接地和编码领域的专家。

在模拟环境辅助生活情境的真实世界设置中（包括处理自然语言的人类请求），该研究的超智能体原型能够结合对机器人的指令，自主组织一系列电磁操控任务。超智能体掌握了与无线通信和感知相关的基础电磁操控技能，并且能够基于人类反馈记住过去经验并从中学习。

据悉，超材料已彻底改变了波的控制方式；在过去二十年中，它们从被动器件，经由可编程器件，发展到了具备传感器的自适应器件，能够实现用户指定的功能。尽管深度学习技术在超材料的逆向设计、测量后处理以及端到端优化中发挥着越来越重要的作用，但其作用终究还是局限于近似特定的数学关系，超材料仍然只是作为人类操作者的代理，实现预定义的功能。

附：英文原文

Title: Electromagnetic metamaterial agent

Author: Hu, Shengguo, Li, Mingyi, Xu, Jiawen, Zhang, Hongrui, Zhang, Shanghang, Cui, Tie Jun, del Hougne, Philipp, Li, Lianlin

Issue&Volume: 2025-01-01

Abstract: Metamaterials have revolutionized wave control; in the last two decades, they evolved from passive devices via programmable devices to sensor-endowed self-adaptive devices realizing a user-specified functionality. Although deep-learning techniques play an increasingly important role in metamaterial inverse design, measurement post-processing and end-to-end optimization, their role is ultimately still limited to approximating specific mathematical relations; the metamaterial is still limited to serving as proxy of a human operator, realizing a predefined functionality. Here, we propose and experimentally prototype a paradigm shift toward a metamaterial agent (coined metaAgent) endowed with reasoning and cognitive capabilities enabling the autonomous planning and successful execution of diverse long-horizon tasks, including electromagnetic (EM) field manipulations and interactions with robots and humans. Leveraging recently released foundation models, metaAgent reasons in high-level natural language, acting upon diverse prompts from an evolving complex environment. Specifically, metaAgent’s cerebrum performs high-level task planning in natural language via a multi-agent discussion mechanism, where agents are domain experts in sensing, planning, grounding, and coding. In response to live environmental feedback within a real-world setting emulating an ambient-assisted living context (including human requests in natural language), our metaAgent prototype self-organizes a hierarchy of EM manipulation tasks in conjunction with commanding a robot. metaAgent masters foundational EM manipulation skills related to wireless communications and sensing, and it memorizes and learns from past experience based on human feedback.

DOI: 10.1038/s41377-024-01678-w

Source: https://www.nature.com/articles/s41377-024-01678-w