Core Insights - The article discusses the advancements in embodied intelligence, particularly focusing on the new training paradigm introduced by Zhujidongli with their LimX DreamActor, which utilizes a multi-data approach for training robots [1][3][25]. Group 1: LimX DreamActor Overview - LimX DreamActor integrates video data, simulation data, and real machine data to enhance robot training efficiency and performance [3][17]. - The training process consists of four steps: data collection using consumer-grade devices, 3D reconstruction with physical parameters, extensive training in simulated environments, and fine-tuning on real machines [7][9][15]. Group 2: Data Utilization Strategy - The multi-data strategy addresses the limitations of each data type: real machine data is expensive, simulation data lacks realism, and video data is challenging to apply due to the absence of physical properties [3][17]. - The approach emphasizes data efficiency, aiming to achieve better performance at lower costs by leveraging diverse data sources [3][16]. Group 3: Technical Innovations - DreamActor employs advanced real machine reinforcement learning (RL) techniques, which significantly enhance learning efficiency and the ability to generalize from simulation to real-world applications [16][18]. - The integration of Real2Sim2Real strategies allows for a more reliable deployment of robots, reducing risks and shortening development cycles [18][20]. Group 4: Historical Context and Evolution - LimX DreamActor is an evolution of previous efforts by Zhujidongli, such as LimX VGM, which focused solely on video data for training robots without real machine samples [21][23]. - The transition from VGM to DreamActor reflects a deeper understanding of data application and the pursuit of optimal data-performance ROI [21][23]. Group 5: Industry Implications - The advancements in the multi-data approach are expected to lower the barriers for participation in embodied intelligence development, enabling more teams to engage in this field [25]. - The article suggests that achieving a balance between efficiency and stability in robot training is crucial for the large-scale application of embodied intelligence [25].

Core Viewpoint - The competition among JD, Meituan, and Alibaba in the humanoid robot and embodied intelligence sectors is intensifying, with each company adopting distinct strategies to secure their positions in this emerging market [1][2]. Group 1: JD's Strategy - JD has made significant investments in humanoid robotics, with three major investments announced on July 21, including Qianxun Intelligent, Zhujidongli, and Zhongqing Robotics, with funding amounts reaching nearly 1 billion [3][4]. - JD's approach combines investment, ecosystem building, and scene application, aiming to create a comprehensive ecosystem that integrates humanoid robots and embodied intelligence [4][9]. - The company has established a dedicated department for embodied intelligence and launched the JoyInside platform to integrate its large model capabilities into various robots and smart hardware [4][8]. Group 2: Meituan's Approach - Meituan has been an early and aggressive investor in humanoid robotics, with investments exceeding 1 billion since 2021, targeting companies like Yushu Technology and Pudu Technology [10][11]. - The company operates two main investment lines: strategic investments through its investment department and financial investments through Meituan Longzhu, focusing on enhancing operational efficiency in its local service ecosystem [13][15]. - Meituan has established its own robotics research institute to deepen its understanding of the industry and improve its technological capabilities [15]. Group 3: Alibaba's Investments - Alibaba has shown strong interest in humanoid robots and embodied intelligence, participating in several funding rounds, including a 300 million RMB investment in Star Motion Epoch [16][18]. - The company has established a subsidiary, Ant Lingbo Technology, dedicated to the research and development of embodied intelligent robots, leveraging its cloud and AI infrastructure [18][19]. - Alibaba's cloud platform has significant capabilities in processing unstructured data, which is essential for training and deploying embodied intelligence models [18][20]. Group 4: Industry Landscape - The competition among major tech giants like JD, Meituan, and Alibaba reflects a broader trend in the Chinese tech industry, with companies like Tencent, Baidu, and ByteDance also investing heavily in humanoid robotics and embodied intelligence [22]. - The market for humanoid robots and embodied intelligence is expected to grow significantly as these technologies transition from experimental phases to practical applications in various sectors [22][23].

Core Viewpoint - The article highlights the recent financing activities of two Shenzhen-based robotics companies, Zhongqing Robotics and Zhujidongli, both of which have received significant investments from major players like JD.com and other notable institutions, indicating a growing interest and competition in the robotics sector in China [1][2][8]. Financing Activities - Zhongqing Robotics completed nearly 1 billion yuan in Pre-A++ and A1 rounds of financing, with key investors including XPeng Motors and JD.com [1]. - Zhujidongli secured a new round of financing led by JD.com, although the specific amount was not disclosed [1]. - JD.com also invested nearly 600 million yuan in another robotics company, Qianxun Intelligent, indicating a strategic push into the robotics field [2]. Product Development - Zhongqing Robotics has launched three humanoid robot products: SA01 for research and education, SE01 as a full-size humanoid robot, and PM01 as a lightweight humanoid robot, along with a quadruped robot showcased in May [4]. - Zhujidongli has released the TRON 1 bipedal robot and plans to launch a full-size humanoid robot in the second half of the year [4]. Application Scenarios - Zhongqing Robotics plans to host a robot combat competition, akin to "WWE," starting in December, and has established partnerships with major companies like NVIDIA and Amazon for diverse applications [4][5]. - Zhujidongli is exploring large-scale applications of its TRON 1 robot in inspection, security, and delivery scenarios [5]. Technological Innovations - Zhongqing Robotics has overcome the Sim2Real technology challenge, creating a unique technical barrier that enhances robot performance in dynamic environments [5]. - Zhujidongli introduced the LimX VGM technology, which utilizes human operation video data for robot operation, emphasizing a multi-source data approach for innovation [5]. Strategic Directions - Zhongqing Robotics aims to accelerate its development pace, increasing production capacity by five times and enhancing R&D investment in embodied intelligence technology [6]. - Zhujidongli focuses on supporting innovation through its general-purpose platform, promoting technological advancements across various sectors [6]. Competitive Landscape - JD.com and Meituan are actively competing in the robotics investment space, with Meituan having invested in 30 robotics-related companies over the past three years [2][9]. - JD.com has established an internal team for embodied intelligence and launched the "JoyInside platform" to attract robotics manufacturers, integrating its AI capabilities into various robotic applications [10].

同样在7月，众擎机器人官宣两轮融资，总金额近10亿元。Pre-A++轮由小鹏汽车作为基石出资人发起设立的星航资本投资，A1轮则由京东领投，溥泉资 本、银泰集团等参投。目前，众擎机器人已具备从本体核心零部件到具身智能、运控算法的全栈自研能力。旗下高兼容性开源人形机器人PM01定价8.8万元 起，已进入量产交付阶段。融资完成后，众擎计划下半年推进规模化试产和交付进程，将生产交付团队规模和能力提升5倍，同时加大具身智能技术的研发 投入。 逐际动力也于7月获得京东战略领投。此次融资将深化逐际动力与京东在零售、物流和服务等领域的协同探索。逐际动力围绕本体硬件设计制造、基于强化 学习的小脑全身运动控制、具身大脑模型的训练策略这三大核心具身智能技术布局，在硬件和具身模型两端同步发力。其发布的LimX VGM利用人类操作视 频数据实现机器人操作的技术创新，全过程零真机样本数据，可实现多平台泛化。全尺寸人形机器人已完成量产准备工作，将于今年下半年公开销售，此前 还发布过双足机器人TRON 1 。 从这一系列投资可以看出，京东在具身智能领域的投资偏好为"硬件+大脑"和"量产能力"。软硬一体的高耦合性，有助于产品在性能和成本上实现 ...

Group 1: Financing and Investment - Two Shenzhen-based robotics companies, Zhongqing Robotics and Zhujidongli, announced recent financing rounds, with Zhongqing raising nearly 1 billion yuan in Pre-A++ and A1 rounds, led by XPeng Motors and JD.com respectively [2][3] - Zhujidongli's latest financing round was strategically led by JD.com, although the specific amount was not disclosed [2][3] - JD.com has also invested in another robotics company, Qianxun Intelligent, which completed a nearly 600 million yuan Pre-A+ round [3] Group 2: Product Development and Offerings - Zhongqing Robotics has launched three humanoid robot products aimed at research and education, including the SA01, SE01, and PM01, and showcased its first quadruped robot, Leopard [2][3] - Zhujidongli has released the TRON 1 bipedal robot and plans to publicly sell a full-sized humanoid robot in the second half of the year [2][3] Group 3: Application and Technology - Zhongqing Robotics plans to host a robot combat competition, akin to "WWE," starting in December, and has formed partnerships with major industry players like NVIDIA and Amazon for diverse applications [4] - Zhujidongli is exploring large-scale applications of its TRON 1 robot in inspection, security, and delivery scenarios [4] - Zhongqing has achieved a technological breakthrough in Sim2Real, enhancing performance in dynamic actions [4][5] Group 4: Strategic Focus and Future Plans - Zhongqing aims to accelerate its development pace, increasing production capacity by five times and investing more in embodied intelligence technology [5][6] - Zhujidongli focuses on supporting innovation in various fields through its general-purpose robotics platform and plans to launch its full-sized humanoid robot soon [6] Group 5: Competitive Landscape - The competition between JD.com and Meituan has expanded into the robotics sector, with Meituan having invested in 30 robotics-related companies over the past three years [7][8] - JD.com is strategically building its capabilities in embodied intelligence, having established a dedicated team and launched the JoyInside platform to attract robotics manufacturers [8]

Core Viewpoint - The article discusses the recent strategic investment by JD.com in LimX Dynamics, aimed at enhancing collaboration in retail, logistics, and services, while promoting the commercialization of embodied intelligence technology [2]. Group 1: Company Overview - LimX Dynamics focuses on supporting and accelerating research, development, and application of a general platform for embodied intelligence robots [2]. - The company employs an "IDS Collaborative Innovation Strategy," targeting innovators, developers, and system integrators as core customers and partners [2]. - LimX Dynamics aims to provide a robust technological foundation for its partners, facilitating the efficient transformation of advanced technologies into practical applications [2][5]. Group 2: Product Development - LimX Dynamics has completed preparations for mass production of its full-size humanoid robot, which will be publicly available for sale in the second half of this year [2]. - The TRON 1 bipedal robot, launched in October 2024, has attracted clients from renowned universities and tech companies, showcasing its capabilities in inspection, security, and delivery applications [4]. - The company has introduced mobile operation, perception expansion, and voice interaction kits for developers, enhancing the TRON 1's industry application potential [4]. Group 3: Technological Innovations - LimX Dynamics released LimX VGM, a technology that utilizes human operation video data for robotic operation innovation, supporting multi-platform generalization without relying on single data sources [4]. - The company emphasizes a product-oriented approach, focusing on efficient model development and addressing actual user needs [5]. - Significant advancements in the Vision-Language-Action (VLA) domain are expected in the second half of the year, enhancing robots' autonomous decision-making and interaction capabilities in complex environments [4].