11月25日，外媒Interestingengineering直接引用极其夸张的标题，报道了一款来自中国的人形机器人 PHYBOT M1 ，内容客观又不吝溢美之词。 视频内容确实相当震撼， 当一个身高近 1.8米、体重近70kg的 " 钢铁巨人 " 纵身跃起，在空中完成流畅的拟 人态后空翻并平稳落地时， 着实令人瞳孔地震。 这款由中国动易科技全栈自研的 产品 ，不仅是全球首款实现该极限动作的全尺寸重型纯电驱人形机器人， 更 以超 10000W瞬时峰值功率的动力表现，标志人形机器人纯电驱时代正式到来，彰显国产核心技术在该领域 的领先地位。 ▍ 全尺寸与高动态性能的协同实现 为 何选择 近 180cm， 70kg的 全尺寸 重型平 台去挑战后空翻？ 机器人大讲堂了解到， 在 PHYBOT M1研 发前，人形机器人领域长期存在全尺寸机型与高动态运动能力难以兼顾的技术瓶颈 。 大多时候， 中小型尺寸机型 是 主流选择 ， 尤其 当前具备奔跑、跳跃等动态运动能力的人形机器人，多采用 中小型尺寸设计。 此类机型因 体惯量 低、重心控制难度小，对关节扭矩及能耗需求较低， 技术实现门槛相 对可控，动易科技此前推出的 PHYB ...

Core Viewpoint - The article highlights the transformative opportunities for traditional inspection methods in the context of advancing intelligent technology, showcasing the development of an intelligent inspection operation platform by Zhongke Shengu, which features autonomous inspection and multi-machine collaboration capabilities [1]. Group 1: Intelligent Inspection Platform - The platform is developed based on the Shengu Open Source Brain System, integrating high-precision motion control and perception decision-making capabilities to create a unified robotic intelligence system [3]. - It enables intelligent scheduling and task allocation among multiple robots, allowing them to autonomously optimize inspection paths and allocate resources in complex scenarios [4]. Group 2: Multi-Modal Perception System - The robotic system employs multi-sensor fusion technology to achieve centimeter-level autonomous navigation in complex indoor and outdoor environments, utilizing 3D laser SLAM and RTK positioning for effective obstacle avoidance and path planning [6]. - The platform includes intelligent upper-level software that consolidates real-time feedback from inspection robots, allowing operators to monitor and manage distributed robots through a user-friendly visual interface [7]. Group 3: Open Architecture and Development Support - The open architecture of the platform supports further development, providing a complete development toolchain and intelligent algorithm library for researchers to enhance functionalities such as mobile operations and environmental interactions [9]. - The system integrates depth cameras and intelligent infrared thermal imaging for superior environmental perception and state recognition under all weather conditions [10]. Group 4: Applications in Industry and Education - The intelligent inspection operation platform demonstrates the practical value of embodied intelligence technology, applicable in industrial parks, smart communities, and research facilities, providing reliable technical support for modern operational management [11]. - It also serves as an efficient platform for teaching and research in fields like robotic engineering, artificial intelligence, and automation, supporting experimental teaching and algorithm validation for researchers [13].

Core Insights - Google’s DeepMind has appointed Aaron Saunders, former CTO of Boston Dynamics, as Vice President of Hardware Engineering, indicating a strategic shift towards robotics [1][18] - DeepMind aims to create an "Android-like" ecosystem in robotics, leveraging the Gemini model to develop a universal AI platform adaptable to various robotic hardware [2][20] Group 1: Aaron Saunders' Background and Experience - Aaron Saunders has over 22 years of experience in robotics, having worked at Boston Dynamics since 2003, where he contributed to projects like BigDog, Cheetah, and Atlas [2][5][9] - He played a pivotal role in the development of Atlas, enhancing its joint degrees of freedom to 28 and enabling complex movements through advanced control algorithms [11][12] - Under his leadership, the Spot robot achieved significant commercial success, being utilized in over 50 countries for various applications [14][18] Group 2: DeepMind's Strategic Goals - DeepMind's strategy focuses on "embodied intelligence," aiming to solve the integration challenges between algorithms and hardware for precise robotic control [22] - The company seeks to establish Gemini as a universal control platform, potentially reducing R&D costs for hardware manufacturers and allowing them to focus on mechanical design and production [27] - This strategic direction reflects a shift from direct ownership of hardware companies to building a foundational technology ecosystem in the robotics sector [25]

Group 1: Company Leadership Changes - On November 25, the Sci-Tech Innovation Board listed company, Senvion New Materials, completed a personnel adjustment with Peng Zhihui, a former Huawei "genius" and co-founder of Zhiyuan Robotics, elected as the chairman of the fourth board of directors [2] - The company has formed a new executive team, with Tian Hua as CEO and legal representative, Zhou Bin as co-CEO and CTO, Zhang Biao as CFO, and Li Yuan as board secretary [2] - This leadership change highlights the capital market's recognition of technology entrepreneurs, with expectations for synergy between Peng's technical background and the company's business [2] Group 2: Industry Developments - Luxshare Precision expects to ship 3,000 humanoid robots this year, having established core capabilities across the entire industry chain except for batteries and some joint modules [3] - The current first-generation production line is semi-automated, with plans to launch a "Manufacturing 2.0" line early next year, featuring high flexibility, adaptability, and automation for complete assembly from parts to whole machines [3] - Sichuan Electric Power has adopted a "drone + robot" combination for the first time to conduct detailed inspections of ultra-high voltage lines, significantly improving efficiency and safety [6] - The new method reduces the workforce from five to three and the time taken from three hours to half an hour, achieving a tenfold increase in efficiency while avoiding high-altitude work risks [6] Group 3: Military and Dual-Use Technologies - The U.S. startup "Foundation" is developing the Phantom MK1 humanoid robot, which may pivot towards military applications, with discussions ongoing with the U.S. Department of Defense for potential uses in aircraft refueling and battlefield scenarios [11] - The company emphasizes its long-term goal of space exploration, but the potential military applications of the robot raise concerns about the blurring lines between civilian and defense technologies [11] Group 4: Standardization in Robotics - A review meeting for the group standard "Technical Requirements for High-Temperature and High-Impact Industrial Robots" was held in Shanghai, aimed at standardizing technical indicators for industrial robots in extreme environments [14] - The meeting was organized by the China Electromechanical Integration Technology Application Association's Intelligent Robotics Branch, with experts from various organizations participating [14]

Core Viewpoint - The emergence of wheeled humanoid robots represents a significant trend in the embodied intelligence robotics sector, indicating a collective preference for this design approach among companies, with over 80% of new products adopting this configuration [1][3]. Group 1: Industry Trends - The current wave of wheeled humanoid robots is characterized by a focus on practical applications, with 45 out of 55 new products featuring a wheeled chassis and humanoid upper body design [1][3]. - The wheeled humanoid robots are seen as a rational choice in the industry, balancing technological feasibility and commercial viability, allowing companies to quickly validate application scenarios and accumulate operational data [3][4]. Group 2: Product Highlights - Several representative wheeled humanoid robots have been launched in 2023, showcasing their application in various sectors such as industrial manufacturing, warehousing logistics, and commercial services [4][5]. - Notable products include: - **Cruzr S2** by UBTECH, featuring a height of 176 cm, capable of carrying 15 kg, and equipped with advanced vision and motion control systems [7]. - **G2** by Zhiyuan Robotics, designed for high-precision tasks in automotive parts production, utilizing reinforcement learning for task optimization [9]. - **Q5** by Xingdong Jiyuan, with 44 degrees of freedom and advanced navigation capabilities for service applications [11][12]. - **ATOM-W** by Dobot, integrating immersive remote operation with high precision for flexible manufacturing needs [15]. - **XMAN-R1** by Qinglang Intelligent, designed for service tasks in high-traffic environments, featuring advanced interaction capabilities [17]. - **DexForce W1 Pro** by Kuawei Intelligent, offering modular development options and high-precision task execution [19]. - **X1-PRO** by Xiangong Intelligent, featuring a flexible design for complex operational environments [22]. - **巡霄** by Youai Zhihui, designed for high-value material handling with a focus on precision [24]. - **ZERITH-H1** by Zero Degree, capable of performing household tasks with advanced perception [26]. - **Quanta X2** by Zivariable Robotics, featuring a high degree of freedom for versatile applications [28].

Core Viewpoint - The article discusses a breakthrough technology called "Adaptive Fission" developed by Qianjue Technology, which significantly enhances the speed and efficiency of robots by mimicking the collaborative functioning of neurons in the human brain [1][8]. Group 1: Why Robots Are Slow - Robots' "brains" operate differently from human brains, leading to slower responses and actions [3]. - Traditional AI models process information step-by-step, resulting in slower decision-making, while brain-like models require multiple iterations for high accuracy, which also slows them down [5][6]. Group 2: Inspiration from the Brain - The human brain achieves quick and precise processing by having neurons work in teams, collaborating to handle complex information efficiently [7]. - Qianjue Technology's "Adaptive Fission" allows key neurons in AI models to split into smaller "assistants," forming an efficient team to process information [8]. Group 3: Three Disruptive Changes from the Technology - The technology ends the era of sacrificing speed for accuracy by dynamically allocating high precision to important tasks and low precision to simpler ones, achieving both speed and intelligence [9]. - It enables "millisecond-level decision-making," drastically reducing the time required for robots to make decisions [10][11]. - The technology is "plug-and-play," allowing already trained robot models to implement this technology without extensive retraining, which is beneficial for the rapidly evolving robotics industry [12][13][14]. Group 4: Performance Metrics - Increasing the number of neurons by 20% can double the reasoning speed, while an 80% increase can quadruple it [15]. Group 5: Future Implications - The breakthrough signifies the arrival of a robot era capable of real-time perception and instant decision-making [17]. - The technology maximizes the potential of existing hardware, suggesting a new era of collaboration between software algorithms and hardware capabilities [18].

Core Insights - Lingxin Qiaoshou (Beijing) Technology Co., Ltd. has completed a series A+ financing round of several hundred million RMB, following its series A round in October, to enhance its product development and commercialization of dexterous hands [1] - The company has achieved a significant milestone by becoming the only global enterprise to mass-produce over a thousand high-degree-of-freedom dexterous hands, capturing over 80% of the global market share in this segment [1] Financing and Investment - The A+ round financing was supported by multiple renowned investment institutions, including Zhejiang Innovation Investment and Dinghui Baifu [1] - The funds will be utilized to support the entire chain from product research and development to capital support for the industrial chain [1] Product Development - The Linker Hand L20, featuring 21 degrees of freedom, is designed to meet both algorithm development and industrial application needs, making it a unique product in the market [2] - The company has introduced the Linker Hand L6 to complete its "Industrial Master" product series, utilizing a self-developed "super electric cylinder" that achieves a driving efficiency of over 90% [2] Market Positioning - Lingxin Qiaoshou has launched the Linker Hand O6, the lightest dexterous hand globally at 370g, significantly reducing end-load and offering a competitive price point of 3999 RMB after subsidies [4] - The company aims to democratize technology and mass production advantages across the industry [4] Technological Innovation - The introduction of the Open TeleDex modular robotic remote operation system allows for a flexible and cost-effective setup for data collection in dexterous operations [5] - The LinkerSkillNet multi-modal data collection system aims to build the largest dexterous operation skill library globally, empowering developers to create new applications quickly [5] Strategic Vision - The company has proposed a new strategy called "Ling Chuang Qiao Jie," extending the concept of "self-manufacturing" to "self-evolution," focusing on creating an ecosystem where robots can continuously learn and evolve [7] - Lingxin Qiaoshou plans to enhance its mass production capabilities, targeting the delivery of 50,000 to 100,000 dexterous hands by 2026 [7]

Core Insights - The article emphasizes that for embodied intelligence to achieve large-scale application, leading chip companies like Intel must overcome challenges in computing architecture [1][3]. Group 1: Challenges in Embodied Intelligence - Recent demonstrations of humanoid robots, such as Tesla's Optimus, have faced criticism for their slow responses and reliance on remote control, highlighting the gap between theoretical capabilities and practical applications [3][4]. - The primary barrier to the deployment of humanoid robots in production environments is the computing power platform, which is currently inadequate for the complex tasks required [4][5]. - The existing humanoid robots typically use a "brain + cerebellum" architecture, where the "brain" handles complex modeling and understanding, while the "cerebellum" manages real-time control tasks [4][5]. Group 2: Computing Power Requirements - The demand for computing power in robotics has increased exponentially due to the integration of action generation models, multi-modal perception, and large model inference [4][5]. - Many companies are resorting to a "two-system" approach, using different chips for the "brain" and "cerebellum," which complicates communication and coordination [4][5]. - The economic aspect of computing power is crucial, as manufacturers need to consider return on investment (ROI) alongside performance metrics like stability, safety, cost, and energy consumption [5]. Group 3: Intel's Solution - Intel proposes a "brain-cerebellum fusion" solution using a single System on Chip (SoC) that integrates CPU, GPU, and NPU, allowing for unified architecture and improved efficiency [6][8]. - The Core Ultra processor achieves approximately 100 TOPS of AI computing power while maintaining similar power consumption levels, enabling faster responses and enhanced privacy [8][9]. - The NPU is designed for lightweight, always-on tasks, ensuring low power consumption and zero-latency experiences, while the CPU has been optimized for traditional visual algorithms and motion planning [9][10]. Group 4: Software Stack and Ecosystem - Intel provides a comprehensive software stack that includes operating systems, drivers, SDKs, and real-time optimizations, allowing developers to start without building from scratch [10][11]. - The oneAPI framework enables seamless integration across CPU, GPU, NPU, and FPGA, facilitating collaboration between existing and new AI hardware [12][13]. - Intel's approach is characterized by openness and flexibility, allowing companies to adapt their systems without being locked into a single vendor's ecosystem [15][16].

Core Viewpoint - The article emphasizes the importance of high-quality, multi-modal, and structured data in advancing humanoid robot technology and its applications, highlighting the donation of the LET dataset to the OpenLoong open-source community as a significant step towards building a unified data infrastructure in the industry [1][11]. Group 1: LET Dataset Overview - The LET dataset, constructed by Leju Intelligent and its partners, is one of the few full-size humanoid robot datasets in China, covering real operational scenarios with over 60,000 minutes of data [2]. - The dataset encompasses diverse task scenarios across three main fields: industrial, commercial retail, and daily life, including six categories such as automotive factories and logistics, with a clear task system comprising 31 tasks and 117 atomic skills [4]. Group 2: Data Collection and Technology Innovation - The dataset records both head and dual-wrist visual streams, providing multi-modal information such as RGB, depth, joint states, and end-effector states, achieving data consistency over 90% through advanced frame grouping technology [5][6]. - Complex tasks are broken down into clearly defined atomic action steps, accompanied by semantic labels, facilitating model understanding of task structures and action logic, thus enhancing behavior understanding and skill learning [7]. Group 3: Building a Trustworthy Data Standard System - The humanoid robot industry faces challenges such as fragmented data sources and inconsistent formats, necessitating a systematic data standard to enhance data quality and model capabilities [12]. - The National and Local Joint Innovation Center for Humanoid Robots has established a comprehensive standard system covering data collection, processing, quality review, and version management, ensuring data quality and usability [14]. Group 4: OpenLoong Community and Future Prospects - The donation of the LET dataset to the OpenLoong community enriches the repository of real-world data, promoting deeper research in task modeling, skill learning, and strategy validation [11][24]. - OpenLoong aims to create a shared data framework to standardize data organization and reuse, with the LET dataset serving as a representative training sample for the industry [20][26].

Group 1 - Figure AI's former product safety chief Robert Grendel has filed a lawsuit claiming that the company's humanoid robots could cause skull fractures and alleges he was fired for raising safety concerns [1] - Grendel warned CEO Brett Adcock and chief engineer Kyle Edelberg about the robots' deadly performance, highlighting an incident where a robot caused a quarter-inch crack on a steel refrigerator door [1] - After submitting documented safety complaints, Grendel was terminated in September, claiming he was asked to prepare a safety roadmap for potential investors, which was later abandoned [1] Group 2 - The emergence of humanoid robots in crowded public events raises significant safety risks due to the lack of comprehensive international or industry standards, leading to compatibility and safety issues [3] - An incident at the 2025 World Humanoid Robot Games involved a robot colliding with an engineer, emphasizing the risks associated with humanoid robots [3] - The insurance industry faces challenges in providing risk coverage for robot companies, including complex risk assessments due to rapid technological advancements and the dynamic nature of risk characteristics [3][4] Group 3 - The lack of long-term risk data in the emerging field of robot insurance complicates accurate risk assessment and premium pricing, as traditional actuarial models are not directly applicable [6] - Insurers are developing innovative financial products to mitigate risks in the robot industry, with major companies like Ping An and China Pacific Insurance launching specialized insurance products for humanoid robots [6][7] - The insurance products available include property insurance, machinery damage insurance, third-party liability insurance, and cybersecurity insurance, aimed at covering various risks associated with robot operations [7] Group 4 - The first breakthrough in robot insurance occurred in early 2022 with the launch of the "Mobile Robot Quality Liability Insurance" by PICC, addressing various product quality issues [8] - In 2023, a pilot project in Changzhou provided insurance for local robot companies, demonstrating regional support for the robot insurance market [8] - By 2025, the insurance industry is transitioning from policy formulation to implementation, with various events and initiatives promoting insurance coverage for robots [10] Group 5 - Local governments are responding to the need for robot insurance, with policies in cities like Ningbo and Hainan promoting the development of humanoid robot application insurance [10][11] - Insurance products are being tailored to cover risks from design defects, operational failures, and cyberattacks, with a focus on third-party liability [11][12] - New insurance offerings include specialized products for exoskeleton robots and smart agricultural robots, indicating a growing market for tailored insurance solutions [12][14] Group 6 - The insurance industry views robot insurance as a means to provide comprehensive risk coverage across the entire lifecycle of humanoid robots, addressing various operational scenarios [15][16] - Insurers are adapting their pricing and claims processes based on accumulated data from existing industrial robots, but face challenges due to the rapid evolution of humanoid robots and the lack of historical data [16]