Core Insights - The article emphasizes the importance of achieving 100% reliability in robotic manipulation tasks through a strategic approach rather than merely increasing data scale [2][4]. Methodology - The proposed methodology consists of three stages: data collection, model training, and real-world inference, which are interconnected and critical for success [2]. - The approach focuses on pattern consistency, model algorithms, and leveraging phase advantages to optimize the transition from perception to action [3]. Pattern Consistency - The article defines the effective action distribution for specific tasks and highlights the need for dynamic alignment among human demonstration, learned strategies, and real-world execution [8][10]. - It identifies potential inconsistencies in traditional imitation learning processes, such as distribution shifts and deployment biases, which can lead to task failures [11][12]. Model Algorithms - The introduction of the Model Arithmetic (MA) method allows for training on new data subsets and merging models without the high costs associated with retraining on full datasets [27][30]. - The MA method successfully integrates different learned manifolds, enhancing model performance beyond that of models trained on full datasets [30]. Phase Advantages - The article discusses the significance of estimating advantage signals directly as a modeling objective, which improves the reliability of state transitions during long-horizon tasks [31][35]. - The proposed Direct+Stage method enhances the stability and smoothness of progress accumulation in robotic tasks [37]. Performance Improvements - Enhanced data collection methods and online strategy recovery trajectories significantly improve model recovery capabilities, leading to higher success rates and reduced retry costs [21]. - The implementation of spatiotemporal enhancements has resulted in increased throughput and task completion rates [23][26]. Conclusion - The article concludes that not all robotic data holds equal value, and the ability to quickly evaluate and select high-quality foundational strategies is crucial for effective research iterations [41]. - The findings suggest that re-evaluating fundamental concepts in reinforcement learning could yield further benefits in robotic manipulation tasks [41].

12月23日，"杭州六小龙"之一的云深处科技启动IPO辅导，辅导券商为中信建投证券。爱企查App显 示，杭州云深处科技股份有限公司已公布上百项专利，专利类型涵盖外观设计、发明专利与实用新型， 涉及人形机器人、足式机器人、无人机等。 ...

Group 1 - The National Robot Industry Index (980022) has increased by 0.46%, with notable stock performances including Lingyi iTech up 3.16% and Nanwang Technology up 2.36% [1] - The robot ETF E Fund (159530), which tracks the National Robot Industry Index, has seen a net inflow of over 260 million in the last two days and over 920 million in the last 20 days, with a total fund size reaching 12.704 billion [1] Group 2 - The first open robot rental platform "Qingtian Rental" was launched in Shanghai, covering over 50 core cities and more than 600 service providers, with rental prices ranging from 200 to over 10,000 yuan [3] - By 2026, the platform is expected to expand its rental services to over 200 cities, transforming the high-threshold robot usage scenarios into a convenient rental model similar to shared power banks [3] - A Morgan Stanley report indicates that while the humanoid robot market will remain hot in 2026, there are risks of hype versus practical value, with challenges in technology development and a potential shakeout of startups [3] Group 3 - The National Robot Industry Index is characterized by a significant weight of humanoid robots, with the top ten weighted stocks accounting for about 40%, including leading companies like Huichuan Technology and iFlytek [4] - The index focuses on high-growth leaders in the sector, indicating strong market competitiveness and high technical barriers [4] Group 4 - The robot ETF E Fund (159530) provides an efficient tool for investors to gain exposure to the entire robot industry chain [5] - There are also off-market fund shares available for the robot ETF, including Link A (020972) and Link C (020973) [6]

新华财经上海12月24日电（记者王鹤）维他动力23日发布无需遥控的消费级具身智能产品——超能机器 狗"大头BoBo"，定价12988元，创始权益价9988元，标志着具身智能向多元消费场景迈出重要一步。 超能机器狗发布后52分钟，订单已突破1000台。产品计划2026年1月启动公测，3月起预定锁单，进入批 量交付阶段。 图为维他动力智能机器狗 （受访者供图） 维他动力介绍，让超能机器狗实现"无需遥控"的，是对真实环境的空间理解能力，即通过强感知、高算 力与自研大模型的深度融合，构建面向真实生活空间的空间智能。 维他动力这款智能机器狗能自主完成跨越全地形的全场景随行、载物、跟拍等多项能力，实现智能伙 伴、全时管家、户外搭子等角色。为了适配更多生活场景，超能机器狗可支持多种外设连接，搭配连接 机械臂、载物栏、相机自拍杆和拖拉小车等，满足出行、记录与协助等需求。 在感知层面，超能机器狗采用多传感器协同架构，实现对复杂室内外环境的稳定感知、空间理解与可靠 避障。在计算与决策层面，搭载128 TOPS高性能端侧AI计算平台，算力水平达到旗舰智能手机的3倍以 上，可同时支持多模态数据的实时处理与大模型推理。 据介绍，"无需 ...

作者丨欧雪 编辑丨袁斯来 硬氪获悉，杭州二白智能科技有限公司（以下简称"二白智能"）近日完成千万级天使轮融资。我们总结了本轮融资信息和该公 司几大亮点： 融资金额及领投机构 融资金额：1000万 融资轮次：天使轮 投资机构：君宸达资本独家投资 资金用途：主要用于二白Mini的量产交付与体验持续优化、长期陪伴能力相关模型与系统研发，以及多形态机器人产品的推进 公司基本信息 成立时间：2019年 公司总部：杭州 核心产品与技术亮点：产品为"二白Mini"随身陪伴机器人，其核心体验并非功能堆砌，而是通过"领养-训练-认主"的深度交互流 程，让每个用户的机器人随着日常相处而成长、变化，形成独特的性格与记忆。公司自研了性格生成算法、情感引擎、稀疏记 忆架构和自然交互技术四大核心技术，旨在实现机器人的初始差异化、后天共生性成长及更自然的互动体验，消除"AI感"。 产品应用：产品定位于日常生活的情感陪伴，覆盖桌面、床头、随身携带等多场景。用户首先在线上"领养"一个独特的数字生 命体并进行训练，待收到实体机器人后通过声纹、人脸识别"认主"，开启长期陪伴。 二白机器人转圈圈演示（图源/企业） 公司旗下"我系二白"账号矩阵运营七 ...

2025年是云深处科技快速发展的一年，业绩实现了快速增长。 本文为IPO早知道原创 作者｜ Stone Jin 微信公众号｜ipozaozhidao 据IPO早知道消息，杭州云深处科技股份有限公司（以下简称"云深处科技"）于2025年12月8日同 中信建投证券签署辅导协议并由后者于12月10日向浙江证监局备案登记。 这意味着，云深处科技成为继宇树科技后、又一家正式启动A股IPO的具身智能企业；同时也是继群 核科技、宇树科技后，"杭州六小龙"中第三家启动IPO的企业。 其中，宇树科技的境内上市辅导工 作已于2025年11月10日全部完成，总计历时4个月。 成立于2017年的云深处科技聚焦具身智能技术创新与行业应用，是专注于四足机器人、人形机器人 及核心零部件的研发、生产、销售和服务的国家级高新技术企业。其坚持自主创新，构建了"感知- 决策-执行"全链路技术体系，其核心的多模态环境感知、跨场景具身导航、AI驱动的感控融合运动 控制等新技术，已在多个新场景中应用落地。 今年以来，云深处科技在产品与技术上持续实现突破。4月，云深处科技重磅发布山猫M20 ——全 球首款专为复杂地形和危险环境设计的行业应用轮足机器人。其 ...

Core Insights - The article discusses the OPENTOUCH framework, which integrates full-hand tactile data collection in real-world environments, addressing the limitations of existing single-modal systems in capturing critical tactile information [3][4][6]. Group 1: Challenges in Tactile Perception - The framework identifies four core challenges in tactile perception: lack of modal information, poor adaptability to real-world environments, difficulties in multi-modal synchronization, and low annotation efficiency [6][7][8][9]. Group 2: Technical Design of OPENTOUCH - OPENTOUCH consists of a three-layer technical loop: hardware perception system, large-scale data collection, and benchmark testing [11]. - The first layer includes a low-cost, robust hardware kit designed for high-precision multi-modal data collection, featuring a full-hand tactile sensing glove and a hand pose tracking glove [12]. - The second layer focuses on building a large-scale multi-modal dataset that covers real-life scenarios, addressing data scarcity [13]. - The third layer establishes a benchmark testing system for cross-modal retrieval and tactile classification tasks, ensuring effective multi-modal integration [15]. Group 3: Performance Validation - OPENTOUCH employs a three-tier validation system to demonstrate its effectiveness, including cross-modal performance, ablation studies, and real-world applications [18]. - The framework shows significant performance improvements in multi-modal fusion models compared to single-modal and linear baselines, with notable metrics in cross-sensory retrieval and tactile classification tasks [20][21]. Group 4: Future Directions and Limitations - While OPENTOUCH represents a breakthrough in full-hand tactile research, there are areas for optimization, such as expanding the tactile dimensions captured, enhancing hardware durability, and improving annotation accuracy in challenging conditions [28][29].

Group 1 - Hangzhou Yundongchu Technology Co., Ltd. officially launched its listing guidance on December 23, 2025, with CITIC Construction Investment Securities as the advisory institution [1] - Yundongchu Technology, one of the "Six Little Dragons" in Hangzhou and a national high-tech enterprise, focuses on the research and development of embodied intelligent robots, establishing a full-link technology system covering perception, decision-making, and execution [1] - The company has completed two rounds of financing exceeding 1 billion yuan in total this year, with over 500 million yuan raised in the C round in December, led by China Merchants International and Huaxia Fund, with participation from funds under China Telecom and China Unicom [1] Group 2 - Huafu Securities believes that humanoid robots can take over repetitive physical labor in human society, representing a significant technological direction for the benefit of humanity [2] - The humanoid robot market in China is expected to have a compound annual growth rate of over 61% from 2024 to 2030, with the market size reaching nearly 38 billion yuan by 2030, and sales projected to grow from approximately 4,000 units to 271,200 units [2] Group 3 - Nandu Property holds a 33% stake in the Saizhi Zhulong Fund, which indirectly holds 1.0638% of Yundongchu Technology [3] - Zhejiang Shuju Culture invested 50 million yuan in 2019 in the Hangzhou Yunqi Venture Capital Partnership, holding a 5.7471% stake, while the Yunqi Fund holds 6.3073% of Yundongchu Technology [3]

Company Performance - iRobot's collapse is being analyzed to understand what went wrong [1] Industry Analysis - The Equity crew is breaking down the reasons behind iRobot's difficulties [1]

"机器人可以让生活变得更好，前提是要先让机器人走进生活，让人们在生活空间中可以自然的和机器 人相处"。近日，消费级具身智能公司Vbot维他动力召开产品发布会，正式推出其首款产品——Vbot超 能机器狗。该公司创始人兼CEO余轶南博士在发布会上的发言表明，这家公司正试图打开家庭机器人应 用的商业化突破口。 Vbot超能机器狗定位"家庭生活伙伴，以12988元定价、9988元创始权益价切入消费市场，凭借"本体智 能+空间智能+Agent智能"三层架构，可实现自主随行、载物、跟拍等全场景能力。值得注意的是，该 产品宣称实现了"无需遥控"的自主行动能力，这在全球消费级机器狗领域尚属首次。 即日起，用户在"Vbot维他动力"官方小程序及"Vbot京东自营旗舰店"上便可参与预订。维他动力表示该 产品将于2026年1月启动公测，3月起进入批量交付阶段。据悉，这款机器狗上线52分钟订单便超过了 1000台，火热程度超出预期。 从生活搭子到社交神器 在社交媒体上，提前尝鲜的媒体人已经展示了Vbot机器狗的各种玩法：让其背着饮料箱一起跟去露营、 拉着坐了小孩的车子在公园溜达、在背上安装摄像头对主人一路跟拍、远程遥控用"狗狗视角" ...