Industry Briefing
A single destination for timely, editor-curated robotics news from around the world.
Chen Pu of Yuan Ke Vision Discusses Transitioning Robot Training Data from 2D to 4D with High Precision
On September 23, Chen Pu, Vice President of Product Development at Yuan Ke Vision, highlighted the importance of high-quality real data for robot training during a seminar in Beijing. He emphasized the need to elevate data from 2D to 4D and improve precision from centimeter to sub-millimeter levels to build a robust 4D data foundation for robotics. This transition is crucial as the industry faces challenges such as weak model generalization and inadequate scene adaptability. Chen noted that current training methods often rely on limited 2D video data, which fails to capture the complexities of three-dimensional space and temporal changes, hindering robots' ability to understand and interact with their environments effectively. Looking ahead, Yuan Ke Vision aims to address these challenges by developing new data collection and training paradigms that enhance dimensionality, enrich modalities, and improve precision. Chen pointed out that while video data serves as a foundation, it is often too simplistic, and the lack of tactile data remains a significant gap in the industry. No further timeline was disclosed at the time of publication.
Robot Training Data 4D Data Collection Industrial Robotics Embodied Intelligence Precision Robotics
Mifengpai Unveils Data Crowdsourcing Initiative for Robot Training Through Daily Activities
On September 23, Mifengpai launched a global initiative to crowdsource data for robot training, showcasing over 50,000 real environments and 5,000 tasks. The event highlighted the need for extensive data on everyday actions, which are crucial for training robots but have not been systematically recorded. Mifengpai's approach combines hardware, an app, and a data engine to facilitate this data collection. This initiative is significant as it addresses the challenge of gathering large-scale data necessary for developing embodied artificial general intelligence (AGI). Mifengpai's infrastructure, including the MEgo collection devices and a user-friendly app, aims to democratize data collection by allowing ordinary users to contribute through standardized tasks. The company has already seen substantial engagement, with 20,000 registered users and over 13,000 data collection tasks submitted in just one month. Looking ahead, Mifengpai has introduced a subsidy plan worth 100 million yuan to support task and equipment subsidies, along with a scene data alliance involving over 50 companies across various sectors. This collaborative effort is expected to enhance the quality and quantity of data available for robot training, ultimately improving robotic capabilities in everyday tasks.
Data Crowdsourcing Robot Training AI Technology Human-Robot Interaction
Shangpin Home Introduces Open-source WorldSimReady-Home Dataset for Robotics Training
Shangpin Home, in collaboration with Tangyuan Technology, has launched the WorldSimReady-Home simulation dataset aimed at addressing the challenges of robotic training in complex home environments. This open-source dataset includes 100,000 square meters of high-fidelity home scenes, 10,000 interactive assets, and 1,000 standardized robotic simulation task examples, allowing for extensive training and testing of various robotic forms. The significance of this initiative lies in its potential to bridge the Sim2Real gap, where robots struggle to perform in real homes despite successful laboratory tests. By providing a diverse range of simulated environments, the dataset enables developers to train robots for navigation, object manipulation, and complex household tasks without the risks associated with real-world trials. Looking ahead, the WorldSimReady-Home dataset represents a foundational step in Shangpin Home's strategy for embodied intelligence. As more teams engage with this open-source initiative, the development of additional datasets for industrial, commercial, and specialized scenarios is anticipated. The effectiveness of this approach will depend on the practical application of the dataset and the successful transfer of learned strategies to real-world settings.
Robotics Training Simulation Data Home Automation AI Digital Twins
Kinetic Blocks Introduces Beta Marketplace for Humanoid Robot Training Data
Kinetic Blocks, a startup based in Oslo, has launched a beta version of a marketplace dedicated to the buying and selling of training data for humanoid robots. This platform became available on September 1, following months of development in collaboration with a select group of data suppliers and early users. The introduction of this marketplace is significant as it aims to streamline the acquisition of training data, which is crucial for the development and enhancement of humanoid robots. By facilitating transactions between data providers and developers, Kinetic Blocks is addressing a vital need in the robotics industry, potentially accelerating advancements in humanoid robot capabilities. Looking ahead, Kinetic Blocks has not disclosed any further timeline for expanding access to the marketplace or additional features. Stakeholders in the robotics sector should monitor this development closely, as it may influence the landscape of humanoid robot training and data utilization.
Computing Humanoids News 1x technologies artificial intelligence egocentric video
NVIDIA and LG Set Ambitious Goal of 100,000 Hours of Robot Training Data by Year-End
NVIDIA and LG Electronics have set a new benchmark in robot training data, aiming for 100,000 hours by year-end. This initiative was announced during a visit by NVIDIA's Senior Director of Omniverse and Robotics Marketing, Min-San Huang, to LG's Yangjae R&D Center in Seoul, following a strategic partnership agreement signed just days earlier. This ambitious target is significant as it surpasses the training data of other companies, such as Ant Group's LingBot-VLA 2.0 model, which has 60,000 hours. The training data will be sourced from a mix of real and synthetic data, leveraging decades of LG's operational data in manufacturing and logistics, enhanced through NVIDIA's Omniverse and Isaac robotics development platform. Looking ahead, LG plans to deploy hundreds of CLOiD robots at the Yangjae data factory, which features various training environments. The company aims to launch a next-generation bipedal robot based on NVIDIA's Isaac GR00T model by Q1 2027. No further timeline was disclosed at the time of publication.
Robot Training AI Robotics Manufacturing Data Analytics
Ropedia Secures $22 Million to Enhance Data Collection for Robotics Training
Ropedia has announced the successful completion of a $22 million pre-Series A funding round, bringing its total funding to $30 million. The investment will be utilized to scale HOMIE, a lightweight, head-mounted device designed to capture first-person human movement and spatial context, which is essential for training robots. This funding is significant as it allows Ropedia to expand its business and technical teams, particularly in hardware, software, and data infrastructure. The company aims to enhance its presence in North America, especially the United States, where most of its clients are located. Ropedia's approach to data collection, which involves generating and structuring data internally, distinguishes it from traditional data-labeling providers. Looking ahead, Ropedia plans to further develop its data platform, incorporating annotation tools and quality analytics. The company is committed to building the necessary data infrastructure for the robotics industry to scale effectively. No further timeline was disclosed at the time of publication.
Artificial Intelligence Artificial Intelligence / Cognition Design / Development Financial Investments News
AI Agents Develop Virtual Environments for Essential Robot Training Data
Robots are becoming more visible in public spaces, captivating onlookers. However, they still lack the versatility needed for tasks in kitchens or factories, primarily due to a significant data bottleneck. Similar to human learning, robots acquire skills through experience, but the process of physically training them in various environments is labor-intensive and time-consuming. This challenge highlights the need for innovative solutions to streamline robot training. By utilizing AI agents to create virtual playgrounds, developers can simulate diverse scenarios, allowing robots to learn efficiently without the constraints of physical environments. This approach could significantly reduce the time and resources required for training, ultimately accelerating the deployment of robots in practical applications. Looking ahead, the development of these virtual training environments may pave the way for more capable robots in various industries. As AI technology continues to evolve, it will be essential to monitor advancements in virtual training methodologies and their impact on robot performance and adaptability. No further timeline was disclosed at the time of publication.
Robotics
Rui Erman Establishes Data Training Facility in Changzhou with 150 Robots Generating 2GB Data Per Minute
Rui Erman has launched a data training facility in Changzhou, equipped with 150 robots capable of generating 2GB of raw data per minute. This facility features two main areas: a basic motion training zone for fundamental robotic actions and a scenario application area designed for diverse operational contexts such as home, industrial, and retail environments. The significance of this initiative lies in addressing critical challenges in the robotics industry, including the scarcity of high-quality real-world data and the high costs associated with data collection. Rui Erman is actively involved in developing standards for data collection, ensuring compliance with the MCAP standard, which emphasizes data accuracy and synchronization. Looking ahead, Rui Erman aims to transition robots from controlled environments to real-world applications, enhancing data collection efficiency. The company is focused on achieving a significant reduction in robot costs to facilitate widespread adoption in households and factories. No further timeline was disclosed at the time of publication.
Data Collection Robotics Machine Learning AI Standards
LG Electronics and Nvidia Aim for 100,000 Hours of Humanoid Robot Training Data
LG Electronics is enhancing its collaboration with Nvidia to expedite the creation of training data for humanoid robots. This initiative follows a memorandum of understanding signed by LG Group Chairman Koo Kwang-mo and Nvidia CEO Jensen Huang, aimed at expanding cooperation in physical AI and mobility. Madison Huang, Nvidia's senior director, visited LG's data factory in Seoul to review the progress of this partnership. The significance of this collaboration lies in its potential to advance the capabilities of humanoid robots through extensive training data. By utilizing LG's CLOiD robots in various simulated environments, including a home setting and a washing machine plant, the companies aim to gather diverse data for training purposes. The data will be processed using Nvidia's advanced robotics solutions, enhancing the learning process for these robots. Looking ahead, LG Electronics plans to fully operationalize the Yangjae data factory by the end of the year, with a target of collecting 100,000 hours of training data. This ambitious goal represents nearly 12 years of continuous operation, marking a significant milestone in the development of humanoid robotics.
All News
MIT Develops SceneSmith: AI Agents Create Virtual Environments for Robot Training
MIT has introduced SceneSmith, a system utilizing AI agents to generate realistic 3D environments for robot training. This innovation addresses the challenge of providing diverse and rich simulation content, which is crucial for robots to learn effectively. By employing a vision-language model, SceneSmith creates detailed indoor scenes that allow robots to practice various tasks before real-world deployment. The significance of SceneSmith lies in its ability to enhance the training process for robots, reducing the time engineers spend on real-world testing. The system constructs scenes with up to six times more objects than previous methods, enabling robots to learn complex skills in a controlled virtual setting. This advancement could lead to more efficient and effective robot training, ultimately accelerating their integration into everyday tasks. Looking ahead, the researchers aim to further refine SceneSmith and explore its applications in diverse robotic tasks. The ability to simulate realistic environments will be critical as robots become more prevalent in various sectors. No further timeline was disclosed at the time of publication.
AI Agents Develop Virtual Environments for Robot Training Using SceneSmith System
AI agents have developed the SceneSmith system, which creates realistic 3D environments such as kitchens and hotels for robot training. This innovative approach allows robots to simulate everyday tasks, enhancing their operational capabilities. The significance of this development lies in its potential to address the skills gap in the manufacturing sector. With over 2 million jobs projected to remain unfilled due to a shortage of skilled workers, effective training solutions like SceneSmith are crucial for preparing the workforce of the future. Looking ahead, the integration of AI in training environments will likely continue to evolve, providing robots with the necessary data to perform complex tasks. No further timeline was disclosed at the time of publication.
MIT and Toyota Develop SceneSmith to Enhance Robot Training with AI-Generated Environments
MIT and the Toyota Research Institute have introduced SceneSmith, a system that utilizes AI agents to create realistic 3D environments for robot training. This innovation addresses the significant challenge of generating diverse simulation content, which is crucial for teaching robots various tasks in a cost-effective manner. The SceneSmith system employs three AI agents, leveraging the advanced vision-language model GPT-5.2, to design intricate indoor scenes. These environments, featuring up to six times more objects than previous methods, allow robots to practice skills in a rich virtual playground, ultimately reducing the need for extensive real-world testing. As the research progresses, the effectiveness of these AI-generated environments will be closely monitored. The team has already demonstrated that robots can successfully navigate and perform tasks in these virtual settings, indicating a promising future for robotic training methodologies. No further timeline was disclosed at the time of publication.
Research Robotics Artificial intelligence Simulation Computer science and technology Machine learning
Shenzhen Luohu District Launches Integrated Humanoid Robot Football Competition and Training Initiative
On September 28, the Luohu District of Shenzhen announced the launch of the first Luohu Cup, an integrated program for humanoid robot football competitions and training, held at Cuiyuan Middle School. This initiative, organized by the Luohu District Education Bureau and supported by Zhiyin Technology and Beijing Accelerated Evolution Technology, aims to enhance AI education in primary and secondary schools through a hands-on, competitive approach. The program emphasizes the importance of embodied intelligence in education, moving beyond theoretical learning to practical applications. By utilizing humanoid robots in football, students will engage in real-world challenges that foster skills in perception, decision-making, and teamwork. The initiative will establish 20 pilot schools, focusing on training over competition, with a goal of developing over 100 certified personnel to support the event. The inaugural Luohu Cup is scheduled to commence in December 2026, with finals and awards in January 2027, leading to the national robotics competition in May 2027. The Luohu District Education Bureau believes that mastering the integration of competition, training, and evaluation will set a new standard for experiential learning in the region.
Humanoid Robots AI Education Robotics Competitions STEM Education
Niantic Spatial Introduces Places Library Featuring 100 Real Environments for Robot Training
Niantic Spatial has launched its Places Library, providing robotics developers with a catalog of 100 real environments for training and evaluation. The assets, available as USDZ files, include two representations for each environment: a Gaussian splat for visual appearance and a mesh for collision detection. This initiative aims to enhance embodied AI training by offering realistic settings that align with gravity and support various simulators, including NVIDIA Isaac Sim. The significance of this launch lies in its potential to improve robot training efficiency. By utilizing real-world environments captured with a standard 360-degree camera, Niantic ensures that robots receive accurate visual and collision data, which is crucial for effective navigation and interaction. The library includes diverse settings, such as medical warehouses and urban streets, allowing developers to test their robots in varied scenarios without the need for extensive in-house data collection. Looking ahead, the Places Library could facilitate more advanced evaluations of robotic behavior in different environments. While the library's impact on performance across the 100 environments remains to be seen, it offers a valuable resource for robotics teams aiming to refine navigation tasks and adapt to changing surroundings. No further timeline was disclosed at the time of publication.
Niantic Spatial Flexion
Landscape Architecture Graduate Transitions to Robot Data Collector Role at Intelligent Company
On September 13, Zhao Yingwei's phone was flooded with messages from classmates and teachers after he was featured in a CCTV interview. Recently hired at an intelligent robotics company, Zhao shared his background in landscape architecture, prompting inquiries on how to enter the robotics field with such a degree. At 25 years old, Zhao's educational journey included a vocational college and a bachelor's degree in landscape architecture. Despite initial enthusiasm, he discovered a bleak job outlook in his field, leading him to seek opportunities in robotics. After applying to several companies, he secured a position as a data collector, where he captures essential data for training robots. Zhao's role involves using a camera-equipped headset and gripper to document everyday tasks, contributing to the development of robotics. He aims to advance within the field by self-learning Linux and Python programming, believing that with determination and available resources, anyone can succeed in this evolving industry. No further timeline was disclosed at the time of publication.
Robotics Data Collection AI Career Transition
Hygon Information Technology Expands AI Chip Focus from Data Centers to Robotics Applications
Hygon Information Technology is preparing to launch a new chip aimed at physical-world applications, including robotics. This marks a significant shift from the company's current focus on data centers, as reported by Chinese media. The new chip, an iteration of the CPU1000 series, is designed to fulfill low-power, embedded, and edge computing needs. It targets various physical AI applications, particularly in robotics, machine vision, and intelligent manufacturing, highlighting the growing demand for AI solutions in these sectors. The launch event is scheduled to take place in Shenzhen, where Hygon will promote its vision of integrating computing power into the physical world. No further timeline was disclosed at the time of publication.
HandEdit Dataset Released: Collaboration with Universities to Advance Robotic Manipulation
Recently, InTime Robotics, in collaboration with Fudan University and Shanghai Jiao Tong University, has officially open-sourced the HandEdit dataset and evaluation benchmark. This initiative focuses on first-person human-to-robot dexterous hand image editing, providing scalable data resources and standardized evaluation criteria for dexterous manipulation learning. The significance of this development lies in addressing the gap between rapid hardware advancements and the insufficient accumulation of robotic operation data. Traditional data collection methods are often costly and time-consuming, limited by the physical structure of different robotic hands. HandEdit aims to convert abundant human operation videos into learnable data for robotic hands, tackling the challenge of structural and functional differences between human and robotic hands. Currently, HandEdit has built over 200 million image editing samples covering 26 URDF configurations, including 13 independent dexterous hands and 13 integrated arm structures. This extensive dataset, along with a unified evaluation system, provides a robust foundation for robotic operation learning and offers a new technical pathway for transforming human operation data into robotic operation data. No further timeline was disclosed at the time of publication.
Robotic Manipulation Data Science Machine Learning AI Robotics
Watney Secures $80 Million to Develop Robots for Data Center Operations
Watney has successfully raised $80 million to enhance its robotic solutions for data centers, focusing on the physical tasks necessary for increasing compute capacity. The funding round, announced on September 17, includes co-leads Valor Atreides AI Fund and Hummingbird Ventures, alongside returning investors such as Conviction, Abstract, A*, and Grant Gordon, bringing total funding to over $100 million. This investment is significant as it addresses labor shortages and operational complexities in critical infrastructure, particularly in AI data centers. Watney's approach emphasizes concrete tasks like last-mile cabling, which allows customers to evaluate performance based on speed, accuracy, and deployment costs. The company aims to provide a pilot program to prospective clients, positioning itself as a deployment-focused business rather than merely a research platform. Looking ahead, Watney's commitment to reliability is noteworthy, claiming over 99.99% reliability and operating the largest fleet of dexterous robots in the U.S. However, the announcement lacks specific metrics on fleet size and task success rates. Future updates will be crucial to understanding how effectively Watney's robots can accelerate data center expansion and improve economic efficiency.
Industrial Automation fundraising
Xiaomi Releases Open-Source Robotics-U0 Model and Training Tools with Significant Speedups
Xiaomi has open-sourced the Xiaomi-Robotics-U0, an autoregressive embodied world foundation model featuring approximately 4 billion parameters and full-scale weight lines of around 38 billion. This release includes training and inference tools designed to enhance robotic applications. The significance of this development lies in Xiaomi's claim of achieving FlashAR+ speedups nearing 83 times, which positions the Robotics-U0 model at the forefront of robot-centric scene, transfer, and video synthesis tasks, as evidenced by its top ranking in WorldArena. Looking ahead, the impact of Xiaomi-Robotics-U0 on the robotics landscape will be noteworthy, particularly in applications requiring advanced scene understanding and video synthesis capabilities. No further timeline was disclosed at the time of publication.
Rhoda AI Evaluates Impact of Web-Video Pretraining on Industrial Robot Performance
Rhoda AI has reported enhancements in industrial manipulation capabilities through scaled web-video pretraining, as detailed in a study released on September 10. The research tested various model sizes, achieving completion rates of 3.7%, 65.0%, 75.3%, and 84.7% in under 100 seconds for tasks like unpacking bearings and sorting waste, with the largest model scoring 94 out of 111. This study is significant as it explores a fundamental aspect of physical AI, demonstrating that larger models, while requiring more computational resources, can lead to improved performance. A separate fixed-size experiment indicated that increasing pretraining compute raised performance from 57.8% to 75.3%, suggesting that the amount of pretraining data and compute plays a crucial role in task execution efficiency. Looking ahead, Rhoda's approach, which utilizes the Direct Video-Action architecture, emphasizes the importance of causal video modeling in robot training. The company’s ongoing evaluations and adaptations of its models will be critical to understanding the future applications of video pretraining in robotics. No further timeline was disclosed at the time of publication.
US rhoda-ai
The Rise of Data Collection in Robotics: UMI and Its Impact on Manipulation Techniques
A new trend in robotics focuses on data collection for manipulation tasks, exemplified by projects like Sunday Robotics' Skill Capture Glove and X Square Robot's TwinDEX. These innovations aim to gather useful manipulation data without needing a complete robot for each demonstration, highlighting the significance of the Universal Manipulation Interface (UMI) introduced in 2024. The UMI's handheld gripper, which incorporates a GoPro, allows users to teach robots skills in everyday settings while leaving the robot behind. This approach emphasizes the importance of early engineering decisions regarding data collection methods, which can significantly influence the robot's learning process and overall performance. Looking ahead, the evolution of data collection tools like UMI and Dobb·E's “The Stick” will continue to shape how robots learn from human demonstrations. No further timeline was disclosed at the time of publication.
Data Collection Sunday Robotics Reward AI XSquare
217 Robots Undergo Training to Improve Elderly Care in Qingdao
In Qingdao, 217 robots are undergoing training at China's first rehabilitation robot training verification center to enhance elderly care. With over 320 million people aged 60 and above in China, and more than 50 million elderly individuals with disabilities, the demand for professional caregivers is significant, with a shortfall of 10 million workers. The training focuses on teaching robots essential tasks such as medication delivery, folding clothes, and assisting with wheelchairs. Each action is broken down into numerous detailed steps to ensure precision, especially when interacting with elderly individuals. The training aims to address the unique needs of seniors, as highlighted by experts who emphasize the importance of personalized care in robotic applications. As the robots learn to perform basic tasks, challenges remain, particularly in understanding dialects and providing appropriate responses to seniors' needs. The Qingdao training center has attracted 45 companies and 210 robot models, indicating a growing interest in the development of care robots. The future of these robots in the market will depend on their ability to interact effectively with humans, rather than just their technological capabilities.
Elderly Care Robots Robotics Training Assistive Technology Healthcare Innovation
The Evolution of Rehabilitation Robots: From Training Tools to Treatment Partners
The WAIC 2026 showcased a significant evolution in rehabilitation robots, highlighting their transition from mere training tools to essential treatment partners. This shift reflects advancements in embodied intelligence, which is becoming increasingly relevant in the field of rehabilitation. This transformation is crucial as it indicates a growing recognition of the importance of human-machine collaboration in therapeutic settings. The integration of rehabilitation robots into treatment protocols can enhance patient outcomes and streamline rehabilitation processes, making them indispensable in modern healthcare. Looking ahead, the focus will be on how these rehabilitation robots will continue to develop and adapt to meet the needs of patients and healthcare providers. No further timeline was disclosed at the time of publication.
Robotics Automation AI
Data Collectors at JD Teach Robots to Live Like Humans Through Daily Activities
At JD's Robot Data Collection Center in Suqian, data collectors are teaching robots to mimic human activities such as cooking and scanning. This innovative approach transforms everyday actions into precise data points, essential for training embodied intelligent models. The center aims to collect over 10 million hours of quality data within two years, recruiting 100,000 full-time and 500,000 part-time data collectors across various environments. This initiative is significant as it bridges the gap between AI and human-like understanding, allowing robots to learn from real-life scenarios. The data collectors, equipped with lightweight devices, meticulously capture actions to ensure the data's accuracy and relevance. Their work exemplifies the evolving relationship between humans and robots, highlighting the importance of human input in AI development. Looking ahead, the center's ambitious goal of extensive data collection will play a crucial role in advancing AI capabilities. As the demand for skilled data collectors grows, this emerging profession is gaining popularity, with experienced collectors earning substantial incomes. No further timeline was disclosed at the time of publication.
AI Data Collection Robotics Human-Robot Interaction
Lan Xiaohuan Discusses China's Data Economy, Robotics, and AI Landscape
Lan Xiaohuan, an economics professor at China Europe International Business School, has authored the bestselling book, How China Works: An Introduction to China’s State-led Economic Development. In his discussions, he highlights the economic factors contributing to China's significant trade surplus and advocates for an enhanced social safety net. Xiaohuan emphasizes the importance of public data infrastructure in shaping the competitive landscape of artificial intelligence, particularly in relation to the United States. His insights reflect the critical role that data plays in driving innovation and economic growth within China. As the conversation around AI and robotics continues to evolve, observers should pay attention to how China's strategies in public data utilization may influence global technological advancements. No further timeline was disclosed at the time of publication.
X Square Launches TwinDEX to Address Data Challenges in Embodied AI Robotics
X Square Robot, based in Shenzhen, has unveiled TwinDEX, a new manipulation platform designed to bridge the data gap in embodied artificial intelligence. This innovative system connects a wearable exoskeleton to a robotic end effector, ensuring identical geometry and functionality, which enhances data fidelity and collection efficiency. The introduction of TwinDEX is significant as it addresses the limitations of traditional data collection methods, which often struggle with the embodiment gap. By utilizing a three-finger, nine-degree-of-freedom architecture, TwinDEX allows for versatile manipulation capabilities while maintaining mechanical reliability and cost-effectiveness. Looking ahead, X Square's approach could redefine how data is gathered for robotic applications, particularly in complex tasks requiring precision. The company has demonstrated TwinDEX's capabilities through a continuous chemistry experiment, showcasing its potential to operate effectively without relying on conventional teleoperation data. No further timeline was disclosed at the time of publication.
Data Collection China XSquare
Orchard Robotics CEO Advocates for Simplified Data Use for Growers
At the Ruggedize ag robotics conference, Orchard Robotics CEO Charlie Wu emphasized that growers should not need to act as data analysts. He highlighted the importance of actionable data, stating that it should facilitate decision-making rather than overwhelm users with numbers. Orchard Robotics offers an AI-powered camera system that captures extensive data on fruit health and growth, processed on-site to accommodate farms with limited connectivity. The company's FruitScope platform allows growers to monitor conditions at a granular level, enhancing operational efficiency. Wu noted that the primary value lies in labor and input savings, with the platform predicting yields with over 95% accuracy. This capability aids in supply chain planning, helping growers optimize labor and resources while making informed decisions about production and marketing. Orchard Robotics has expanded its focus from apples to various crops, including grapes and cherries, and is exploring new markets internationally. The company currently operates hundreds of systems across tens of thousands of acres, aiming for broader adoption. Wu stated that growers can expect a return on investment of three to ten times in their first year by leveraging the platform effectively.
Agtech Artificial intelligence Deeptech Precision agriculture Startups & funding US & Canada
Kinetic Blocks Launches Marketplace for Humanoid Training Data Acquisition
On September 1, Oslo-based startup Kinetic Blocks introduced a gated beta of a marketplace tailored for the buying and selling of humanoid training data. This platform aims to streamline the traditionally slow and complex process of dataset procurement by replacing bilateral licensing deals with standardized commercial transactions. The significance of Kinetic Blocks' launch lies in its potential to address the challenges of physical data acquisition in the embodied AI sector. By allowing data suppliers to list various datasets, including egocentric human video and teleoperation recordings, the platform seeks to establish clear market values and mitigate the opaque rights management that has historically plagued robot learning data procurement. Looking ahead, Kinetic Blocks plans to expand its engineering and commercial teams in the coming months while preparing to open a seed funding round in the fourth quarter of 2026. This development comes amid a competitive landscape where foundational model developers are increasingly seeking innovative strategies for sourcing real-world telemetry data.
Data Collection Kinetic Blocks Dataset Europe
The Limitations of Robotics and Automation Without Proper Data Management
Manufacturers are increasingly investing in robotics and automation, yet many fail to address the foundational issue of manual data management. This oversight significantly limits the return on investment (ROI) for new technologies. As automation does not inherently solve workforce challenges, the focus must shift to training and upskilling employees to enhance productivity in the automotive and machine-building sectors. The gap between automation investments and effective data management is critical, as it can stifle the potential benefits of new technologies. Without a robust data foundation, companies may find their automation efforts falling short of expectations, leading to inefficiencies and missed opportunities for growth. This situation highlights the importance of integrating data management strategies alongside automation initiatives. Looking ahead, companies must prioritize the development of comprehensive data management systems to fully leverage their investments in robotics and automation. No further timeline was disclosed at the time of publication, but the ongoing skills gap in the workforce remains a pressing concern for manufacturers aiming to optimize their operations and achieve sustainable growth.
Mountain Club Launches Humanoid Robot Training Program for Young Learners
On August 15, Mountain Club officially launched its first training course in collaboration with Lion Technology and Jumping Equation. This program integrates theoretical learning, hands-on practice, and competition experience, centered around the second World Humanoid Robot Sports Competition's dance project. Over ten days, 14 students embarked on a journey from understanding to practical application of humanoid robotics. The course began with an introduction to robots, guided by instructor Zhang Hengxi, covering competition rules and robot structures. Students engaged in hands-on activities, disassembling and reassembling robots, and progressed to AI programming, communication, and decision-making. They learned to control robots through code, understand sensor functions, and explore visual systems, transforming abstract concepts into functioning robots. From classroom learning to real-world applications, students visited the competition site from August 22 to 24, observing how humanoid robots perform under pressure. This experience provided them with a comprehensive understanding of the robotics industry, emphasizing the long journey from laboratory development to market readiness. The training concluded, but the lessons learned extend beyond technical skills, highlighting the importance of practical experience in robotics education.
Humanoid Robots Robotics Education AI Programming Robotics Competitions
Data Collection Challenges in Robotics: Addressing Physical Limitations and Gaps
Since the beginning of the year, major companies and startups have significantly increased their investment in data collection. Companies like Qianxun Intelligent, Lingqiao Intelligent, and Lingchu Intelligent have announced ambitious targets for collecting millions of hours of data. Meanwhile, Guanglun Intelligent has completed a financing round of 1 billion yuan, becoming the world's first embodied data unicorn. JD.com has unveiled a comprehensive infrastructure for embodied intelligent data collection, planning to mobilize 600,000 people for crowdsourced data gathering across 64 training sites in 27 cities. This surge in data collection efforts highlights the industry's focus on building data sets, annotation teams, and simulation environments. However, a critical physical limitation is being overlooked: most teams simplify data collection to perception-level image and point cloud gathering, neglecting the essential motion data from the interaction between robots and physical environments. According to the Guizhou Provincial Big Data Bureau, only 500,000 hours of compliant data from real physical interactions currently exist in China, while the China Electromechanical Integration Technology Application Association estimates that commercializing robotics requires at least tens of millions of hours of data support, indicating a gap exceeding 99% based on a conservative estimate of 10 million hours. The current challenges in real-world data collection stem from structural constraints that create a physical ceiling. While virtual environments can generate training data at low cost, the gap between simulation and reality is widening as model complexity increases. The AI Index Report 2026 from Stanford HAI reveals that robot manipulation success rates drop from 89.4% in simulated environments to just 12% in real home settings. This discrepancy underscores the need for real physical interaction data, as many robots struggle in unstructured environments like stairs and uneven surfaces, which are crucial for embodied intelligence applications. Continuous data collection is necessary for iterative algorithm development, yet many data collection vehicles are designed for specific scenarios, leading to high costs and inefficiencies in cross-environment deployments.
Data Collection Embodied Intelligence Robotics Simulation Physical Interaction
Index Launches as the Largest and Most Diverse Robot Training Dataset Ever Created
Today, Figure has unveiled Index, a groundbreaking robot training dataset designed to address the data scarcity for general-purpose robots. Over the past four months, the company has developed a unique pipeline to collect real-world physical data, achieving over 264,000 app downloads across 108 countries and 44,000 weekly active users contributing to the dataset. The significance of Index lies in its ability to provide diverse and high-quality data essential for training AI systems like Helix. With over 16 million videos uploaded and 30 minutes of video processed every second, the dataset captures a wide range of tasks, objects, and environments. Figure has committed to investing over $1 billion in data and compute resources over the next year to further enhance this initiative. Looking ahead, Figure aims to scale its data collection efforts significantly, with plans to increase the dataset's capabilities and diversity. The company is already witnessing promising generalization results from its AI stack, Helix, and will share more insights on its findings in the near future. No further timeline was disclosed at the time of publication.
robotics AI data collection machine learning technology
Humanoid Robot Crashes During Training for World Humanoid Robot Games in Beijing
A humanoid robot experienced a severe crash during a training sprint in Beijing, colliding with a cushioned wall and nearly splitting in half. This incident occurred as part of the preparations for the World Humanoid Robot Games, also known as the 'Robot Olympics,' set to begin on August 22. The event will feature 2,056 machines from 666 teams across 16 countries. The crash, which was captured on video and has garnered over 13 million views online, raises questions about the robot's programming and perception capabilities. Experts note that such failures are crucial for testing robots under challenging conditions, allowing developers to gather data to enhance performance and safety before real-world deployment. As the World Humanoid Robot Games approaches, with 30 competitive events planned, the increase in participation—up 138 percent in teams and quadrupled robot entries—highlights the growing interest and advancements in humanoid robotics. No further timeline was disclosed at the time of publication.
AI and Robotics
Micro1 Achieves $500 Million Gross Run Rate Amid Surge in AI Training Data Demand
Micro1, a data-labeling startup, has seen its gross annual run rate increase from $100 million to $500 million in just eight months, driven by the high demand for unique AI training data. The company retains about 60% to 70% of this figure, resulting in a net annual run rate between $150 million and $200 million. This significant growth highlights the robust market for AI training data, with Micro1's revenue trajectory indicating a strong demand that can support multiple players in the sector. While competitors like Mercor and Handshake have surpassed Micro1 in gross revenue, the startup's expansion reflects a broader trend in AI spending, which may soon rival expenditures on computing resources. Looking ahead, Micro1 is poised for continued growth as it increases contract sizes and expands its synthetic data generation capabilities. The company is also navigating controversies regarding the sale of off-the-shelf data, particularly concerning its stance on not selling to Chinese AI developers, as articulated by founder Ali Ansari. No further timeline was disclosed at the time of publication.
AI Startups data labeling micro1 reinforcement learning
Xiaomi Unveils New-Generation Humanoid Robot After Four Months of Auto-Factory Training
On August 19, 2026, Xiaomi introduced its new-generation humanoid robot at the World Robot Conference, following four months of training in an auto-factory environment. This robot, named CyberOne 'Tieda', stands 1.70 meters tall and weighs 66 kilograms, featuring 66 degrees of freedom. The significance of this development lies in the robot's ability to perform florist interactions autonomously, without relying on preset scripts. This capability is driven by advanced model-based autonomous decision-making, showcasing Xiaomi's commitment to innovation in robotics. Looking ahead, industry observers will be keen to see how Xiaomi's humanoid robot integrates into various applications and the potential impact on the robotics market. No further timeline was disclosed at the time of publication.
51World Unveils New Data Collection Tools to Enhance Embodied AI for Robots
51World, a Beijing-based technology company, has introduced a new suite of data-collection devices aimed at overcoming the critical shortage of high-quality training data for embodied AI systems. CEO Li Yi emphasized that the lack of precise data is a significant barrier to developing stable and capable humanoid robots. The newly launched AperEgo headset features a multi-camera system and synchronized sensors to capture a comprehensive view of the environment, while additional devices for the wrist and fingers enhance data collection on hand movements. This integrated hardware and software approach is expected to improve data accuracy and efficiency significantly. Looking forward, 51World aims to enhance the efficiency of embodied AI in data collection and training. The Chinese humanoid robot market is projected to reach 15 billion yuan (approximately US$2.2 billion) by 2026, with significant growth anticipated in 2027 as production and applications expand. No further timeline was disclosed at the time of publication.
Hexagon Robotics Begins Training AEON Humanoid Robots at Schaeffler's Facility in Germany
Hexagon Robotics and Schaeffler have initiated the training of AEON humanoid robots at Schaeffler’s Humanoid Gym in Germany. This marks a significant step towards deploying at least 1,000 AEON robots in the coming years, utilizing a Train-Validate-Deploy model tailored for industrial environments. The collaboration is crucial as it allows both companies to enhance AEON's industrial capabilities while building Schaeffler's expertise in operating and integrating humanoids into their production processes. The training will focus on imitation learning and refining robot policies to ensure reliable performance in real manufacturing applications. Looking ahead, the training at the Humanoid Gym is expected to expedite AEON's deployment across various manufacturing workflows within Schaeffler over the next six months. This initiative not only supports the integration of humanoids into Schaeffler's operations but also aims to strengthen overall manufacturing performance and scalability in automation.
Components Computing News Training aeon factory automation
A3 and Mahoning County Launch Free Robotics Training in Ohio with $499,000 Grant
The Association for Advancing Automation (A3) and the Mahoning County Career & Technical Center (MCCTC) are set to provide free robotics and automation training in Ohio, supported by a $499,000 workforce grant. This initiative aims to enhance the skills of approximately 3,000 residents by June 30, 2027, through the Ohio Individual Microcredential Assistance Program (IMAP). This program is significant as it addresses the growing demand for skilled workers in technology-driven careers, particularly in robotics and automation. A3 will offer online courses covering essential topics such as industrial robotics and robot safety, which are crucial for small and medium-sized manufacturers looking to adopt these technologies effectively. Looking ahead, the program will continue to expand, with five additional courses planned for release in the coming months. Ohio residents interested in these opportunities can enroll through the A3 website, contributing to a stronger workforce prepared for the future of manufacturing and automation in the region.
News Robotics a3 advanced manufacturing automate automation training
A3 and Mahoning County Career & Technical Center Launch Free Robotics Training Initiative in Ohio
The Association for Advancing Automation (A3) and the Mahoning County Career & Technical Center (MCCTC) have announced a new initiative to provide no-cost robotics and automation training throughout Ohio. This program is supported by a $499,000 workforce grant aimed at enhancing the skill sets of residents in the state. This initiative is significant as it addresses the growing demand for skilled workers in technology-driven careers, particularly in robotics and automation. By offering free training, A3 and MCCTC are making it easier for individuals to gain valuable skills that are increasingly sought after in the job market. Looking ahead, stakeholders will be monitoring the impact of this training program on workforce development in Ohio. The success of this initiative could serve as a model for similar programs in other regions, promoting the importance of skills training in adapting to technological advancements. No further timeline was disclosed at the time of publication.
Trossen Robotics Collaborates with Stereolabs to Enhance Physical AI Data Collection
Trossen Robotics has announced a partnership with Stereolabs to integrate high-fidelity stereo vision into its Physical AI platforms. The collaboration features the Stereolabs ZED X Mini scene camera and dual ZED X Nano wrist cameras, providing synchronized, training-grade visual data for robot-learning teams. This integration is significant as it enhances Trossen's offerings in the Physical AI sector, allowing for improved data collection and analysis. The inclusion of advanced stereo cameras is expected to elevate the capabilities of Trossen's hardware suite, which includes the Trossen Workbench and Rivet platforms designed for bimanual manipulation. Looking ahead, the collaboration aims to streamline the development of robot learning applications by providing robust visual data. No further timeline was disclosed at the time of publication.
Generalist Leverages Human Demonstration Data for Enhanced Robot Learning
Generalist, a robotics startup valued at $2 billion, utilizes human demonstration data to train robots on real-world tasks. Developed through collaboration among Toyota Research Institute, Columbia University, and Stanford University, the Universal Manipulation Interface (UMI) enables the collection of training data via puppet-like end effectors and GoPro cameras. This innovative approach allows collaborative robots to learn tasks such as washing dishes and picking up objects more efficiently. The significance of Generalist's work lies in its ability to create adaptable robots that can recover from errors in real-time, a feature demonstrated at the Automate event. The company showcased its models performing various tasks with Universal Robots and Flexiv arms, highlighting the intelligence of these systems in handling unexpected challenges. This capability has the potential to reshape perceptions of automation in industrial settings. Looking ahead, Generalist aims to further refine its models to maintain a competitive edge in a rapidly evolving market that has seen over $4 billion in investments. The company’s commitment to developing versatile robotic solutions across diverse applications will be crucial for its growth and adoption in the industry. No further timeline was disclosed at the time of publication.
Academia / Research Arms / Manipulators Artificial Intelligence Artificial Intelligence / Cognition Assembly Cobot Arms
Shelford Group Expands START Programme for Robotic Surgery Training in South East England
The Shelford Group has announced a significant expansion of its Surgical Training in Advanced Robotic Technology (START) programme. This initiative will extend accredited robotic surgery training to surgical trainees in the South East of England, starting from the 2026/27 academic year. Previously available in the North East, North West, and East of England, the START programme will now include regions such as Thames Valley, Wessex, and Kent, Surrey, and Sussex. This expansion aims to enhance the skills of surgical trainees in robotic surgery, which is becoming increasingly important in modern medical practices. As the demand for advanced surgical techniques grows, the expansion of the START programme is a crucial step in ensuring that more surgical trainees receive high-quality training. Stakeholders will be watching closely to see how this initiative impacts the quality of surgical care in the newly included regions. No further timeline was disclosed at the time of publication.
58.com Partners with Woan Robotics to Enhance Robot Training in Real Homes
In August, Woan Robotics signed a strategic cooperation agreement with 58.com’s subsidiary, Xingxing Kexing Technology. This partnership aims to bridge the gap between AI-driven home robots and real-world living scenarios. Woan Robotics' AI brain, OneModel, requires practical household experiences to function effectively, which 58.com can provide through its extensive local service platform. The collaboration will initially focus on health and commercial environments, with plans to expand into real home applications. With over 90 countries served and more than 5 million households impacted by Woan's products, the partnership is set to enhance the post-sale service network for robots, utilizing 58.com’s talent pool for maintenance and support. Future developments will explore human-robot collaboration, using real-life job processes from 58.com’s platform as training material for robots. This innovative approach positions 58.com not just as an information intermediary but as a supplier of training data for robots, potentially reducing error rates in household robots by leveraging real-world practice before deployment.
Home Robotics AI Training Robot Maintenance Human-Robot Collaboration
Joint Robot Training in Guiyang Achieves Success with Over 2.04 Million Views
On August 7, 2026, a training program on the safety and effectiveness of domestic joint surgery robots was successfully held at Guizhou Provincial People's Hospital. This event was organized by the Sichuan International Medical Exchange Promotion Association and featured experts from various prestigious hospitals, including Sichuan University West China Hospital and the PLA General Hospital. The training included surgical demonstrations and academic discussions, showcasing the capabilities of Yuanhua Intelligent Technology's surgical robots. The significance of this training lies in its demonstration of advanced robotic technology in orthopedic surgeries, particularly in total knee arthroplasty (TKA) and total hip arthroplasty (THA). Yuanhua Intelligent's robots, equipped with precise navigation capabilities, received high praise from attending experts. The event highlighted the importance of integrating robotics into surgical practices, which can enhance surgical precision and improve patient outcomes. Looking ahead, the continued development and application of robotic-assisted surgeries will be crucial in the orthopedic field. Experts shared insights on the future of robotic surgery, emphasizing the need for ongoing training and knowledge sharing among medical professionals. No further timeline was disclosed at the time of publication.
Orthopedic Robotics Surgical Training Medical Technology Healthcare Innovation
Chengdu's Robot Training Facility Prepares Robots for Real-World Applications
Chengdu's new robot training facility, located in the W7 building of the Chengdu Science and Technology Innovation Island, has commenced trial operations. The facility features various training zones focused on electronic skin, home services, industrial operations, retail services, and rehabilitation, creating a comprehensive hardware system that includes robots, mechanical arms, and sensory devices. This initiative is significant as it addresses the limitations of traditional laboratory training by simulating real-world scenarios. The facility's design ensures that data collected during training reflects practical applications, which is crucial for the future development of robots capable of gentle handling and safe interactions. The training center not only supports local enterprises but also extends its services to robotics research teams across the province. Looking ahead, the facility aims to enhance the integration of artificial intelligence in everyday life. As robots learn tasks such as cash handling in simulated environments, they move closer to becoming integral parts of our daily routines. The opening of this training school marks a pivotal step towards advancing embodied intelligence from mere mobility to functional autonomy.
Robotics Training AI Applications Industrial Automation Data Collection Smart Home Technology
Addressing the Challenge of Tactile Interaction Data in Robotics Development
The global robotics community is facing a significant challenge: the lack of real physical interaction data, particularly tactile data. While visual datasets and first-person videos are becoming increasingly common, the industry struggles to gather the nuanced feedback from tactile interactions with various materials. Current tactile data collection methods either involve expensive laboratory-grade sensors or rely on simulations that do not accurately reflect real-world physics. Recognizing this challenge, Handzhi Innovation has developed a new approach that balances cost and effective data collection. Founded by prominent figures in robotics research, including Academician Liu Sheng and Dr. Li Miao, the company aims to transform high-precision tactile perception technology into scalable infrastructure. Their HANDX series tactile gloves exemplify this effort, integrating up to 800 high-precision tactile points in a lightweight, flexible design that supports dual-mode transmission and offers significant operational capabilities. The industry is at a crossroads, as it lacks standardized, low-cost, scalable tactile data collection infrastructure. Handzhi Innovation's strategy involves democratizing tactile data collection through widespread use of their gloves in real-world scenarios, significantly reducing marginal costs and increasing data diversity. The accompanying software platform provides a comprehensive toolchain for data management, making it easier for users to engage in tactile data collection without extensive development efforts.
Tactile Data Collection Robotics Technology Data Infrastructure AI Machine Learning
Shanghai Concludes Spring Training for Robotics Teams with Real-World Testing
Thirty-two maker teams showcased their robots in a real-world environment, tackling tasks such as CityWalk, coffee delivery, and urban inspections. Despite their capabilities, the robots faced challenges, including reliance on remote control and limitations in long-distance tasks. This testing highlighted the need for standardized data systems among teams to facilitate collaboration between carbon-based and silicon-based entities. The training culminated in a short film depicting a robotic dog navigating various tasks, revealing both its limitations and the technological advancements made by participating teams. The film emphasized the importance of human oversight and decision-making in enhancing robotic functionality, such as modifying elevator access for robots. Shanghai has introduced the 'Dual-Base Friendly Agreement' to ensure that silicon-based entities do not compromise human safety or privacy. The team is drafting guidelines for creating friendly communities that integrate these technologies. As robots transition into everyday life, it is crucial to align technological advancements with urban planning, social norms, and human psychology.
Robotics Urban Navigation Human-Robot Interaction AI Technology
Jiangsu's AI Innovations Showcase Rapid Robotics Training and Practical Applications
Jiangsu is demonstrating the capabilities of AI through various practical applications, including a quadruped robot conducting inspections and individuals mastering laser welding in just three days. This was highlighted during the 'Vibrant China Research Tour' organized by the Publicity Department, where over 100 journalists explored AI practices across eight cities in Jiangsu. The 'AI Mirror' ecosystem in Nanjing features an exoskeleton device that enhances user strength and a development center that connects product innovation with market testing. Since its launch in November, the center has engaged over 200 companies and facilitated nearly 70 million yuan in transactions, showcasing a successful model of technology integration. In Wuxi, various robots are being trained for precision tasks, while Nantong is transforming traditional manufacturing with AI systems. Suzhou's collaboration with local chip manufacturers is accelerating the laser industry. Jiangsu's diverse AI applications are providing concrete answers to the fundamental question of what technology can achieve in today's era.
AI Robotics Manufacturing Laser Technology
Virtuix Partners with Tesla to Enhance Robot Training Amid Stock Market Challenges
On July 27, Virtuix Holdings announced that Tesla has purchased its first Omni One Enterprise system for the Optimus humanoid robot division. This marks Virtuix's first public enterprise-level collaboration with Tesla, expanding its business from consumer entertainment and defense medical to industrial robotics. The significance of this partnership lies in its potential to address the current challenges faced by the Tesla Optimus project, particularly in teaching robots to operate in real environments. Virtuix's system allows operators to remotely control robots in virtual settings, collecting valuable human operation data for AI training while simulating complex scenarios safely. Looking ahead, the collaboration could pave the way for Virtuix to attract more industrial clients, despite the initial market reaction that saw its stock drop over 30%. No further timeline was disclosed at the time of publication.
Robot Training Virtual Reality Industrial Robotics AI Development
Encord and Zander Labs Use EEG Data to Advance Physical AI Training Techniques
In a warehouse in San Leandro, California, a worker is participating in an experiment that combines a data collection helmet with EEG sensors to train robots. This collaboration between Encord and Zander Labs aims to address the scarcity of real-world training data for physical AI, which is a significant challenge in the field. The importance of this experiment lies in its potential to generate valuable training data by capturing the neural activity of operators during tasks. This data can inform robot models about operator states, such as confusion or focus, enabling more efficient training and resource allocation. Encord is also collecting remote control data and first-person videos to create a comprehensive data production system. Looking ahead, the integration of EEG helmets, muscle sensors, and detailed annotations could revolutionize how robots are trained, providing the necessary real-world data that is currently lacking. No further timeline was disclosed at the time of publication.
Physical AI Robot Training Data Collection EEG TechnologyRobotics needs a service framework.
RSF defines a common language for robot service capability, lifecycle operations, certification pathways, and service-provider networks.
Join the RobotToday community on LinkedIn
Daily robotics news, in-depth analysis, conference highlights, and discussions with professionals worldwide.
