手臂还没抬起来，外骨骼就知道你要干啥了 ！ 美国 佐治亚理工学院等机构 的研究团队 近日公布了 一套 "读心"外骨骼系统。这 套系统 在你动作发生前就 预测你的意图， 500-550毫秒内给你加个buff 。 传统的外骨骼系统往往存在诸多限制：要么只能按照预设程序运行，缺乏实时感知能力；要么体积庞大、布线 复杂，难以在日常生活中使用；更关键的是，大多数系统无法准确预测使用者的运动意图。 佐治亚理工团队的这套系统巧妙地解决了这些痛点。 首先是传感器的突破 。研究团队开发了一种超薄柔性 EMG（肌电图）传感器，可以像创可贴一样贴在皮肤 上。这种传感器采用了 蛇形金纳米膜电极 设计，即使在 30%的拉伸应变下，经过300次循环测试后电阻变化 依然微乎其微。 怎么做到的？研究人员给外骨骼装上了柔性生物电子传感器，专门 捕捉你的肌肉信号 。再配合云端深度学习 算法，实时分析你想干啥。 与商用的硬质凝胶电极相比，这种柔性干电极不仅 避免了长时间佩戴造成的皮肤刺激 ，还能与不规则的人体 表面实现贴合，最大程度减少了运动伪影。在对比测试中，柔性传感器获得的 EMG信号质量与商用传感器完 全相当，信噪比表现相近。 简单来说，就 ...

Core Insights - The article discusses the challenges in end-to-end robot learning, particularly focusing on the "Sim2Real" gap, which is primarily caused by the inadequacy of simulation environments to accurately replicate real-world scenarios [1][6][10]. Group 1: Challenges in Robot Simulation - Current simulation environments struggle with three main issues: insufficient realism in replicating real-world scenarios, high costs in scene asset acquisition and system configuration, and time-consuming data collection processes [1][5]. - The core obstacle is the performance drop during the Sim2Real transfer, which stems from the fundamental differences between simulated and real-world environments, such as object appearance, lighting effects, and spatial geometry [1][6]. Group 2: Existing Simulation Frameworks - Various simulation frameworks have been developed, but none meet the three critical requirements: high visual fidelity, accurate physical interaction, and efficient parallel scalability [3][6]. - Traditional simulators often compromise on either visual realism or physical accuracy, leading to ineffective training for robots [6][7]. Group 3: DISCOVERSE Framework - DISCOVERSE is an open-source simulation framework developed by Tsinghua University in collaboration with other institutions, integrating 3D Gaussian Splatting (3DGS), MuJoCo physics engine, and control interfaces into a unified architecture [5][10]. - The framework aims to bridge the Sim2Real gap by enhancing the realism of simulations through a three-layer innovation approach, focusing on accurate digital representation of real-world scenes and objects [10][12]. Group 4: Performance and Efficiency - DISCOVERSE significantly improves simulation speed, achieving rendering speeds up to 650 FPS on high-end hardware, which is three times faster than competing solutions [19][20]. - The framework supports a wide range of asset formats and robot models, enhancing compatibility and reducing the need for extensive configuration [21][22]. Group 5: Testing and Results - In comparative tests, DISCOVERSE outperformed other mainstream simulators in zero-shot transfer success rates across various tasks, demonstrating its effectiveness in real-world applications [24][27]. - The framework also enhances data collection efficiency, reducing the time required to gather demonstration data from 146 minutes to just 1.5 minutes, thus accelerating algorithm iteration [29]. Group 6: Future Implications - DISCOVERSE is positioned as a versatile robot simulation framework capable of supporting various complex tasks, with potential applications in robotics, drones, and autonomous driving sensors [30]. - The release of the framework's code and API aims to facilitate adoption by developers and enterprises, marking a significant milestone in the robotics industry [30].

Core Viewpoint - The article discusses the transition of the robotics industry from "automation" to "intelligence," highlighting the role of Professor Li Zhijun and his development of the insightOS, a domestically produced intelligent operating system for robots, aimed at enhancing their autonomous decision-making capabilities [1][4][19]. Group 1: InsightOS Development - The insightOS is designed to address the complexity and flexibility challenges in robot applications, integrating intelligence deeply into the system rather than treating it as an external module [4][5]. - The system aims to achieve robot autonomy by embedding AI capabilities such as perception, reasoning, and planning into the operating system's core services [4][5]. - A unified semantic understanding layer allows non-experts to command robots using natural language, significantly lowering deployment barriers and enabling large-scale collaboration among intelligent agents [5][8]. Group 2: Dynamic Adaptation and Flexibility - InsightOS features dynamic adaptability through built-in elastic scheduling and multi-model collaboration, enabling real-time resource allocation based on complex scene requirements [8]. - The system can autonomously identify anomalies and plan recovery strategies, ensuring stability in dynamic environments, thus addressing the critical challenge of making robots not just "usable" but "effective" [8][9]. Group 3: Ecosystem Construction - The company is building a comprehensive ecosystem that connects application scenarios, hardware adaptation, and algorithm integration, aiming to create a universal system tailored to domestic industry needs [9][10]. - InsightOS has achieved hardware decoupling and capability atomization, allowing compatibility with various robot types, including industrial arms and drones, by abstracting hardware differences into a unified "semantic behavior" [9][10]. - The insight Studio development platform supports rapid integration of third-party algorithm models, significantly reducing the barriers and costs for robot manufacturers in deploying and tuning algorithms [10][12]. Group 4: Vision and Mission - InsightOS aims to create a "self-controllable operating system" for Chinese robots, addressing the reliance on foreign platforms and mitigating data security risks [13][15]. - The system is built on decades of research from Harbin Institute of Technology, ensuring a strong foundation in both operating systems and robotics [15][18]. - The company plans to establish a complete product matrix and has already deployed insightOS in several domestic humanoid robots, receiving industry recognition [18][19].

1、 人形机器人首登十五运会开幕式，奏响千年青铜句鑃 11月9日，第十五届全国运动会（以下简称"十五运会"）开幕式在广州奥林匹克体育中心举行，现场，随 着第一声青铜颤音传出，3台来自深圳企业优必选的全自主具身智能人形机器人，代表粤港澳三地，作为 全球首个人形机器人开幕嘉宾，敲响着由广州南越王墓出土的8件战国青铜句鑃仿制品，正式拉开了十五 运会开幕式的大幕，这创造了两项突破性纪录，既是国家级综合性运动会首次引入人形机器人作为开幕嘉 宾，也是全球首次人形机器人奏响千年青铜礼乐。青铜句鑃仿制品作为古代礼乐文明的重要载体，音色清 越、韵律古朴，演奏需精准控制敲击位置与力度才能呈现丰富旋律，即便对人类乐师而言也极具挑战。 2、 首届中国（国际）机器人辩论大赛决出冠军，智能思辨新高度 11月9日，首届中国（国际）机器人辩论大赛决赛在北京经开区举行，来自北京、上海、广州、天津等全 国各地学校和企业的14支团队参赛。经过初赛、复赛和决赛的激烈角逐，最终松延动力"小诺队"获得冠 军。据悉，本次赛事填补了国内机器人"智能思辨"类赛事空白，推动人工智能技术在自然语言处理、逻辑 推理及情感交互领域的深度探索，搭建机器人技术交流平台， ...

Core Insights - Tesla plans to rapidly increase production capacity, with a focus on the mass production of the Optimus humanoid robot, expected to launch the Gen3 version in 2026 [1][2] - The hybrid architecture of the Optimus robot, which combines linear actuators and rotary actuators, shows significant potential for commercial applications in various industrial tasks [1][2] Production and Market Position - Tesla's production line in Fremont aims for an annual output of 1 million units, while a Texas facility targets 10 million units [1] - Kepler Robotics has already launched the K2 Bumblebee humanoid robot, priced at 248,000 yuan, achieving mass production ahead of Tesla [2][4] Technical Advancements - Kepler's K2 Bumblebee utilizes a similar hybrid architecture to Tesla's Optimus, overcoming challenges in kinematic modeling and joint torque control [4][5] - The K2 Bumblebee features a self-developed planetary roller screw actuator, achieving an energy conversion efficiency of 81.3% and a continuous operation time of 8 hours on a 1-hour charge [10] Software and Ecosystem Development - Kepler Robotics emphasizes a self-developed software ecosystem, enhancing the K2 Bumblebee's capabilities in intelligent interaction and task execution [11][13] - The Kepler Studio platform allows developers to create complex industrial task scenarios with a user-friendly interface, lowering the technical barrier for programming [14] Strategic Collaborations and Funding - Kepler has established partnerships with key component manufacturers and automotive companies, creating a comprehensive support system for technology iteration and mass production [16] - The company has completed three rounds of financing in six months, gaining recognition from seven A-share listed companies in the automotive and robotics sectors [16] Future Outlook - The market potential for humanoid robots is significant, with Tesla's Optimus expected to reduce production costs to around $20,000 when scaled to millions of units [19] - Kepler Robotics is positioned as a leading brand in the hybrid architecture humanoid robot sector, with ongoing efforts in hardware and software integration for commercial viability [19]

Core Viewpoint - 2023 is recognized as the "Year of Large Models," while 2025 is anticipated to be the eve of the explosion of "World Models," which are reshaping the core logic of embodied intelligence and driving the evolution of the robotics industry towards higher-level intelligence with environmental cognition and proactive decision-making [1]. Summary by Sections World Model Definition and Characteristics - The World Model represents a significant advancement over traditional robotic frameworks, which follow a linear "perception-decision-control" chain. It enables robots to understand, predict, and plan by creating a high-dimensional cognitive model of the real world, allowing for proactive reasoning rather than merely executing commands [2][4]. - The World Model's capabilities are characterized by three internalization features: spatial internalization (transforming 2D data into 3D semantic space), rule internalization (learning basic physical rules), and temporal internalization (integrating historical and real-time data for continuous understanding) [3]. Development and Application of World Models - The concept of World Models has evolved over three decades, beginning with Richard S. Sutton's Dyna algorithm in 1990, which integrated learning, planning, and reaction mechanisms. This laid the theoretical groundwork for its application in robotics [7]. - The transition to practical applications began in 2018 with the publication of the "World Models" paper, which demonstrated the potential of World Models in complex dynamic environments through deep learning techniques [9]. - Since 2019, advancements in computational power and multimodal technologies have accelerated the development of World Models, leading to their integration into real-world applications, such as Tesla's Full Self-Driving (FSD) system and Xiaopeng Motors' training environments [10]. Impact on the Robotics Industry - The industrialization of World Models addresses key challenges in traditional robotics, such as data scarcity and high training costs. For instance, World Models can generate vast amounts of virtual scenarios from minimal real data, significantly reducing training expenses [12]. - World Models enable large-scale training scenarios, allowing for comprehensive testing across diverse conditions, which enhances safety and reliability in robotics applications [13][15]. - The cognitive leap provided by World Models allows robots to make human-like decisions, improving their adaptability in complex environments and expanding their application value [15]. Challenges in Industrialization - Despite the potential of World Models, challenges remain, including the need for improved memory and generalization capabilities to handle long-duration tasks in complex environments [16]. - There are still fundamental differences between simulation and reality, particularly in aspects like texture, dynamic consistency, and non-deterministic events, which can affect performance during real-world deployment [18]. - Ethical considerations, such as decision-making transparency and data privacy, are critical as the complexity of World Models increases [18]. Future Trends - The integration of World Models with multimodal technologies is expected to enhance robots' environmental understanding and predictive capabilities, leading to more reliable and generalized performance [19]. - The evolution towards end-to-end solutions centered around World Models will reduce reliance on manual rules and high-precision maps, streamlining development processes [21]. - The shift towards a cloud-edge collaborative computing architecture will facilitate large-scale scenario simulations and model training, optimizing performance and reducing deployment costs [21]. Conclusion - The development of World Models marks a transformative shift in the robotics industry, addressing traditional challenges and redefining the technological landscape. By 2030, the market for robots equipped with World Models is projected to exceed 3 trillion yuan, with significant contributions from various sectors [22].

人才市场的热度是产业活力最直接的晴雨表。猎聘最新发布的《 2025三季度人才供需洞察报告》（以下简称 《报告》） 显示， 2025年三季度具身智能领域新发职位同比猛增72.86%，这一增速不仅远超同期科技行业 平均水平，更比人工智能行业整体增速高出近18个百分点，形成 明显的 人才需求 "虹吸效应"。 | | | | 新发职位同比增长排名 | 三级职能 | 2025Q3新发职位同比增长 | | --- | --- | --- | | No.1 | 机器人工程师 | 88.14% | | No.2 | 汽车设计工程师 | 80.31% | | No.3 | 车身工程师 | 75.93% | | No.4 | 飞行器设计与制造 | 74.48% | | No.5 | 保险销售 | 70.06% | | No.6 | 投资/理财顾问 | 69.23% | | No.7 | 行政后勤/总务 | 67.45% | | No.8 | 算法工程师 | 66.37% | | No.9 | 汽车产品规划 | 65.78% | | No.10 | 药品研发 | 65.36% | | No.11 | 装配工程师 | 57.99% ...

近日，一款名为 Aceii One 的 AI 网球机器人在海外众筹平台 Kickstarter 引发关注， 筹集资金已突破 76 万 美元 ，成为体育科技领域的一匹黑马。这款集人工智能、机器人技术和网球训练于一体的创新产品，正在重 新定义网球训练的方式 。 Aceii One 的智能化体现在其感知与决策系统 。 它搭载了两个由人工智能驱动的立体摄像头，如同机器的 "眼睛"，持续追踪球员在球场上的实时位置。 ▍ 便携与移动： Aceii One 如何改变训练体验 传统的网球发球机往往笨重且固定， 而 Aceii One 的设计首先解决了便携性问题 。 它由一个可装载 120 颗 球的球斗和一台球机组成，组装后的大小接近一个高尔夫球袋或登机箱。当需要运输时，两者可以分离，轻松 放入大多数汽车的后备箱。这意味着球友可以更方便地将它带到任何球场，不再受设备搬运的困扰。 更关键的是它的移动能力。与固定不动的传统发球机不同， Aceii One 底部配备了由两个电机驱动的四个脚 轮，让它能在球场上自由移动。官方数据显示，它能在 1.2 秒内以 3.5 米 / 秒的加速度覆盖半个球场。这种 移动性是其被称为"机器人"而非" ...

11月8日， 以 "与湖州共未来"为主题的2025湖州具身智能机器人领域产业投资对接会在风景秀丽的西塞科学谷成功举办 。 本次大会汇聚了来自具身智能机器人领 域的顶尖学者、行业领袖、头部投资机构及产业链上下游企业代表约 110人共襄盛会，深入探讨 具身智能 产业最新发展动态和未来趋势。 大会 议程设置精准务实，紧紧围绕技术前沿、产业融合与生态构建三大核心维度展开，通过高密度的思想碰撞与战略发布，为与会嘉宾描绘了一幅清晰的湖州具身 智能发展蓝图。 其 标志着湖州在抢抓未来产业新赛道、构建人工智能创新高地上迈出了坚实一步。 湖州 市委副书记、市长连坤明 ， 湖州市 副市长金凯 ， 市政府秘书长沈斌章 ， 市商务局党委书记、局长陈少鹏 ，市科协党组书记、常务副主席顾云飞，市产业 集团党委书记、董事长沈志伟，以及湖州有关部门、各区县负责人，产业链企业、应用方、投资机构共同出席了本次会议。 大会开幕式上 ，多位学术界权威专家发表 了 致辞。中国自动化学会机器人专业委员会副主任陶波 指出，具身智能正从实验室走向工厂，融入到生产中，真正参与 到我们物理世界的改造和提升 ；清华大学教授、智能技术与系统国家重点实验室副主任孙富 ...

Core Insights - The year 2025 marks a dual explosion period for embodied intelligence, with the government report listing it as a key future industry to cultivate, and global technology routes forming a three-layer architecture of "body + brain + cerebellum" [1] - The second Zhongguancun Embodied Intelligence Robot Application Competition attracted over 150 teams, showcasing innovations from both enterprises and top research institutions [1] - The upcoming conference on November 19, 2025, aims to bridge the gap between laboratory innovations and market applications [3][4] Event Highlights - The conference will feature keynotes from industry leaders and experts, focusing on the empowerment of industry through embodied intelligence [6][10] - A roundtable forum will discuss the transformation from competition to market, addressing real-world needs and resource bottlenecks for small teams [10][11] - The afternoon session will focus on cutting-edge research and technological breakthroughs, addressing core technical bottlenecks in the field [11][12] Industry Challenges - The performance of embodied intelligence robots relies heavily on key components such as servo drives and tactile sensors, with some still dependent on imports, posing a challenge for industry development [12] - The conference will explore paths for domestic innovation and industry collaboration to strengthen the foundation of the embodied intelligence sector [12] Future Outlook - The event is positioned as a critical node for insights into trends, resource connections, and breakthroughs in the embodied intelligence industry [13]