Pollen Robotics (part of Hugging Face) has launched a Hoverboard two-wheel accessory for Reachy Mini, enabling Reachy Mini Wireless to move around and explore, and allowing users to drive it remotely with virtual joysticks via an iOS/Android mobile app, control its head movements, and view live video.
Pollen Robotics released a tutorial demonstrating how to install the NFC Accessories Kit in the head of Reachy Mini Wireless, enabling magnetic accessory detection with just four screws and no glue. Once installed, the robot can automatically recognize NFC accessories placed on its head, and each accessory can unlock a different personality in the Reachy Mini conversation app or trigger custom actions. The kit also includes extra NFC tags and magnets for users to create their own accessories and AI interactions.
Pollen Robotics has released the Reachy Mini Hoverboard, a compact two-wheeled mobile base that lets Reachy Mini Wireless roll and explore its surroundings.
Pollen Robotics releases the Reachy Mini Hoverboard, a compact two-wheeled mobile base that turns the Reachy Mini Wireless into a mobile robot.
CCTV Finance's "Economy Half-Hour" aired a feature on Daimon Robotics' young team, whose core members include post-95 CTO Du Yipai and post-00s product managers and algorithm engineers; 80% of the company's staff are post-90s and 40% are post-00s. Daimon builds a full-chain capability spanning "perception-data-models-deployment applications" around vision-tactile sensors, with the goal of enabling robots to sense weight, direction, and contact state after grasping. Daimon has also launched the TouchSpark innovation research program, which long-term accepts project proposals from scholars and lab teams at universities and research institutes worldwide and provides funding and research resource support.
On September 16, 2026, Nanjing University and Daimon Robotics signed and unveiled the "Nanjing University-Daimon Robotics Embodied Intelligence University-Enterprise Joint Laboratory" at Nanjing University's Suzhou campus. It is the first embodied intelligence university-enterprise joint laboratory of Nanjing University's School of Intelligent Science and Technology. The two parties will conduct joint research on directions such as robotic manipulation and tactile manipulation models, and will combine Daimon's accumulated work in tactile hardware, real-world data, and engineering validation to advance scenario validation and the translation of results. The laboratory will also carry out joint graduate training, technical personnel training, and regular academic exchanges.
Daimon Robotics was recently named to both the EY 2026 "Most Promising Seed Enterprise" list and the 2026 Forbes Asia "100 to Watch" list. The EY selection has been held 16 times in Greater China, and more than 60 companies from past lists have gone public. This year's Forbes list, themed "AI-driven innovation," selected 100 companies from 16 countries and regions across Asia-Pacific. Forbes noted that Daimon is filling the physical interaction capability gap needed for robots to advance toward general intelligence.
At IROS 2026, held September 27 to October 1 in Pittsburgh, USA, Daimon Robotics presented its full-chain physical intelligence system spanning perception, data, and models under the theme From Touch to Intelligence, centered on the Daimon-TWM tactile-anchored world model.
Topstar has launched a humanoid dual-arm robot solution for sorting irregular parts. Operators issue commands in natural language, and TOP Agent parses the tasks and coordinates visual perception, motion control, and force-controlled gripping, cutting changeover time from hours to minutes. On the hardware side, the humanoid dual arms have a rated payload of 8 kg and repeat positioning accuracy of ±0.05 mm, are equipped with a binocular vision system and a force-controlled adaptive gripping module, and support 24/7 continuous operation. The solution has completed deployment validation in the supply chain of a leading 3C brand and can optionally be equipped with edge computing and industrial large models to extend to processes such as appearance inspection.
WeRide disclosed that its subsidiary JingShuo Technology has received a Pre-A round of tens of millions of USD co-invested by Jiayu Capital, the HKSTP Venture Fund, Sirius Capital, and others, marking the first external capital introduced since its founding; the round is still in progress. JingShuo Technology builds data and model infrastructure with its WorldEngine data platform and self-developed world model GENESIS-Robotics, and currently has 100+ customers and ecosystem partners, with order backlog in the billions of yuan and annual revenue growth exceeding 100%.
RealMan Intelligent is advancing the real-world deployment of robots through three metrics: working hours, human intervention rate, and human efficiency ratio. Its integrated joint modules have passed CR L3-level reliability certification and achieve a mean time between failures of 50,000 hours, and nearly 100 robots are already providing continuous service in industrial and commercial scenarios.
Standard Robots will exhibit at the 2026 World Internet of Things Expo, held from October 30 to November 1, 2026, at the Wuxi Taihu International Expo Center, at booth A3-09. The company will exchange views with visitors on the application of robots in real industrial scenarios and explore further deployment possibilities.
Swancor Advanced Materials reported revenue of 1.209 billion yuan in the first three quarters, with net profit attributable to shareholders from its main business of 63.5181 million yuan, up 4.91% year on year. R&D spending reached 384 million yuan, up 1,182.01% year on year, of which 360 million yuan went to robotics R&D.
The State Administration for Market Regulation released the first batch of 6 national standards for humanoid robots, effective April 2027. Dobot, as a drafting participant, focused on two core standards: motion control and task operation. Dobot said it will use real engineering data from production lines to formulate capability descriptions and verification requirements in the standards.
Qiyuan Robot has integrated with Tencent WorkBuddy. Users can issue instructions in natural language, and WorkBuddy interprets the intent and breaks down the tasks, which Qiyuan Robot executes using its environmental perception and motion capabilities, forming an interaction chain from task understanding to robot execution. Tian Hua, CEO of SWANCOR and Qiyuan Robot, shared this progress on the cooperation at the Tencent Global Digital Ecosystem Summit.
Galileo has released an upgraded petrochemical intelligent inspection solution, centered on the new-generation Ex2 intelligent bionic legged explosion-proof robot and paired with an embodied AI inspection platform. The Ex2 has obtained IIB+H₂ extended explosion-proof certification and reaches the T5 temperature class; it can climb 35° slopes and 30cm steps, with runtime upgraded to 7 hours and payload upgraded to 100 kilograms.
Apex Intelligence has released and open-sourced its first large model, Apex-35B-V1.0, with 35 billion total parameters, about 3 billion parameters activated per token, and support for a 262,000-token context; it is based on Qwen3.6-35B-A3B and optimized through mid-training and post-training.
A small team has built a model that generates photorealistic simulation renders from G-buffers at about 2 FPS, aimed at robotics and physical AI. The team is surveying practitioners' simulation stacks (Isaac Sim/Lab, MuJoCo, Gazebo, Genesis, Unreal/Unity, or in-house), their environment and digital twin setup workflows, and whether their own robots reproduce well in simulation. Participants can get 1 month of free use of the team's product.
PHYMI officially released its first robot, Sagi, on October 28. It adopts a "four-wheeled-legged humanoid configuration" combining four wheeled legs with a humanoid upper body, dual arms, and a rear box, targeting scenarios such as office delivery, community errands, and outdoor camping transport.
Why it matters: The first four-wheeled-legged humanoid robot presents a complete design approach covering configuration trade-offs, a three-layer model, and a data system, allowing comparison with existing bipedal routes to understand the choice of form.
On October 23, the 10th 'Two Weeks, One Chain' Greater Bay Area robot full-industry-chain supply-demand matchmaking event, hosted by the Artificial Intelligence (Robotics) Administration of Longgang District, Shenzhen, and co-organized by the Shenzhen Robotics Association, was held at the Longgang District Robot 6S Store, with the theme of robot exterior components and intelligent manufacturing. Yunhai Zhidong proposed a brushless motor solution for the cargo compartment opening of delivery robots, shortening opening time from 6 seconds to 3 seconds, with noise below 60 decibels and cost reduced by 15%; Dongguan Huanya Model used vacuum casting and low-pressure injection to achieve small-batch manufacturing at the hundred-unit level, shortening delivery cycles by more than 70% and reducing costs by 40%-60%.
Shenzhen LaborX Technology Co., Ltd. (LaborX) is a full-stack industrial embodied intelligence company wholly incubated by LK Group, building a three-part in-house system of the LaborX OS industrial system, a wheeled mobile manipulator robot, and a high-precision force-controlled dexterous hand.
The Shenzhen Longgang District AI (Robotics) Administration has released the 2026 Longgang District robotics application scenario project list and opened challenge applications. The application period runs from 9:00 on October 29, 2026 to 18:00 on November 30. The list is based on real business needs proposed by the issuing units, and seeks challengers from legal entities, private non-enterprise units, or industry associations nationwide. Scenarios are in principle to be implemented within Longgang District, and project construction periods are in principle no more than 12 months, with a maximum of no more than 2 years.
On October 26, the first demo day of the Beijing Academy of Artificial Intelligence (BAAI) "Source Innovation Program" was held at the Zhongguancun International Technology Trading Center. Five projects were unveiled: Linjie Core Technology, Lingshu Qizhi, Zhangliang Chang Technology, Cangqiong Jizhi, and Qianxiangliang Technology, covering directions including 3D self-selecting V-RRAM chips, neuromorphic embodied intelligence, Hamiltonian world models, physical intelligence in extreme environments, and unified world models of surface physical environments.
The Robot Carnival is now open for recruitment, seeking ideas and works from the robotics field so that robots' creativity can be seen by the world. The event is announced by the Moon Explorer Embodied Intelligence Community; follow that community for participation details and event specifics.
Intel will hold an Embodied Intelligence and Robotics Frontier Trends Exchange on November 10 in Zhangjiang, Shanghai, limited to 100 attendees, showcasing a multi-channel GMSL real-time perception solution based on the Intel Core Ultra X7 358H platform and a 358H embodied intelligence complete system solution.
In the 50 days since GPT-6 Astra was released, the embodied AI industry has conducted intensive evaluations and discussions around it, and general-purpose large models have begun directly taking over robot operation. Tests by SenseTime and several institutions concluded that Astra is basically capable of completing embodied tasks, but still has a clear gap from practical use in real environments, underperforming in millimeter-level contact, dynamic control, and response speed. The article also reviews industry uses such as using Astra directly for policies, reviewing VLA actions, building simulation environments, and data annotation, as well as chain reactions including cooling financing and major companies stepping up data collection.
Why it matters: It reviews evaluation results, company reactions, and industrial chain changes in the embodied AI industry within 50 days of GPT-6 Astra's release, presenting the differentiated landscape after general-purpose large models entered robot operation.
The article reviews the commercialization paths of over 20 underwater robotics companies, noting that their core is selling work orders rather than hardware, with customers paying based on sea trial passage and operational results. Shihang Intelligence completed an A round of over 1 billion yuan in June 2026, with orders exceeding 1 billion yuan in the first half of the year, and launched a full-ocean-depth platform from 0 to 10,000 meters and the registered "Cangqiong" ocean embodied model. The article analyzes three tracks—work-class ROVs, ship cleaning, and scientific research inspection—and mentions incremental space in Middle East and African markets with a compound annual growth rate exceeding 14%.
Unitree Technology released its Q3 2026 financial report: third-quarter operating revenue was 665 million yuan, up 69.93% year on year; net profit attributable to shareholders was 113 million yuan, down 17.37% year on year. Revenue for the first three quarters was 1.817 billion yuan, up 55.71% year on year; net profit attributable to shareholders was 387 million yuan, up 270.54% year on year; non-recurring net profit was 349 million yuan, down 18.78% year on year.
At the 2026 Agibot Innovation Day launch event, Agibot disclosed that its revenue for the first half of 2026 was close to 2 billion yuan, and cumulative output of its general-purpose AI robots reached 20,000 units. The event also claimed that Agibot holds a 43% share of the global humanoid robot market. For comparison, Unitree Technology disclosed the previous day that its revenue for the first three quarters of 2026 was 1.817 billion yuan, up 55.71% year on year, with net profit attributable to shareholders of 387 million yuan, up 270.54% year on year.
Unitree Technology disclosed its 2026 third-quarter report on the evening of October 28. Third-quarter revenue was 665 million yuan, up 69.93% year on year; revenue for the first three quarters was 1.817 billion yuan, up 55.71% year on year. Q3 net profit attributable to shareholders was 113 million yuan, down year on year; the company said this was mainly due to increased share-based payment, R&D, and operating investment. R&D investment in the period was 127 million yuan, nearly 20% of revenue.
Tesla plans to use Optimus humanoid robots to plug and unplug charging cables for Robotaxi as a transitional solution before wireless charging is deployed. Currently, Robotaxi still requires staff to charge manually at Supercharger stations; the Cybercab supports only Supercharging and has a hidden NACS port inside its rear bumper. Tesla has already planned a Robotaxi charging hub in Austin equipped with 80 wireless charging units. Optimus has just started early production at the Fremont factory, with first deliveries to enterprise customers planned for 2027.
LimX Dynamics' Beijing company opened a new site on October 28 at the Zhongguancun (Haidian) Embodied Intelligence Innovation Park, and the Beijing team will take on more core R&D tasks. The Beijing company will serve as a strategic anchor in the north, focusing on building three centers: embodied brain R&D, talent and industry-academia-research, and industrial ecosystem collaboration. It will continue to upgrade the humanoid brain system COSA and embodied VLA capabilities, refine fully autonomous long-horizon tasks, and ensure real-world scenario validation. The park has already gathered more than 50 companies across the industrial chain, and LimX Dynamics, as one of the first "Digital-Intelligent Sharing" entitlement companies, will leverage park resources to accelerate technology iteration and scenario validation.
The Yufeng 250-L LiDAR Edition quadcopter drone kit is aimed at research and education. It consists of three core modules: a flight controller, an onboard computer, and LiDAR. It can deploy SLAM, planning, and control algorithms to achieve real-time navigation in unknown environments.
At the 2026 World IoT Expo (Wuxi), Galaxea is presenting three major embodied applications: a healthcare scenario deployment in partnership with Kunshu Xingqi, a robot front warehouse with on-site ordering and multi-robot collaborative autonomous picking and packing, and floor exercise champion Kengo taking on high-difficulty moves. The expo runs from October 30 to November 1 at Wuxi Taihu International Expo Center, Hall A3-03.
At the 23rd Pharmaceutical Industry Ecosystem Development Conference, Atomrobot shared "The Intelligent Leap of AI + Robotics in the Pharmaceutical Industry." Wu Jiangfeng, general manager of the delta robot division, explained the practical value of AI vision, delta robots, SCARA robots, and automated production lines in processes such as material transfer, sorting, loading and unloading, and inspection and rejection. The company says it has accumulated over a thousand implemented projects in the pharmaceutical field and provides delta robots, SCARA robots, visual inspection, and flexible automated workstations for scenarios including solid preparations, sterile preparations, and pharmaceutical packaging materials.
Qiyuan robots have been integrated with Tencent WorkBuddy. Users issue commands in natural language; WorkBuddy understands the intent and breaks down the task, while Qiyuan robots perceive the environment and execute it, forming an interaction chain from task understanding to robot execution.
RoboCanvas, a visual tool aimed at ROS 2 beginners, has released a free public beta that supports visually designing and configuring robots, validating designs, and generating corresponding ROS 2 project files. The author says the goal is to help students and hobbyists avoid detours with URDF, launch files, dependencies, and simulation environment setup, but the tool will not replace learning ROS 2, and feedback from people with ROS 2 experience is still being sought.
RoboCanvas has released a free public beta that supports visually designing and configuring robots, validating designs, and generating corresponding ROS 2 project files. The author says the goal is to lower the barrier to entry for ROS 2, URDF, launch files, and simulation environments, mainly for students and hobbyists, and stresses that the tool will not replace learning ROS 2.
REEMAN Robotics is developing a commercial autonomous tennis ball-collecting robot for professional tennis academies, multi-court clubs, and commercial tennis facilities, with the goal of reducing repetitive ball-collecting work and allowing coaching teams to focus on training. The product is in the development and technical evaluation stage. Development priorities include a ball-storage target of about 200–300 balls, autonomous navigation, tennis ball detection and collection, automatic charging, and court surface compatibility; these are engineering goals rather than verified mass-production specifications. REEMAN has not yet announced a confirmed launch date, final price, mass-production specifications, or sales channels.
The embodied AI sector has fallen into a race of prioritizing fundraising and demos and serving narratives rather than real deployment; projects that can be stably commercialized account for less than 5% of all humanoid robot projects. The article points out that a demo-first evaluation system leads to prototype rework, stalled pilots, and project delays, and argues for shifting assessment metrics to continuous operating hours, fault reproduction rate, environmental generalization capability, and unit maintenance cost, and for first refining closed loops in clearly bounded scenarios such as factories, warehousing, and inspection.