Skip to content
#Perception and localization
Jun 3Wed
  1. Stial/斯帝尔科技28

    STIER Uses Physical AI Closed-Loop Technology to Enable a New Model for Flexible Grinding and Efficient Polishing of Bent Pipes

    STIER has built an integrated AI flexible grinding solution covering perception, analysis, decision-making, and execution, targeting the difficult grinding and polishing of special-shaped bent pipes such as liquid-cooling bent pipes and exhaust bent pipes. Its self-developed SeeV perception system combines 3D imaging to identify contours and deviations; the NextBrain™ AI Mushroom Cloud system plans trajectories and enables one-click changeover; and the six-axis AI flexible grinding robot is equipped with the PolishX flexible force control system, achieving contact force control accuracy of ±0.5% FS and supporting small-batch, multi-model customized production.

May 30Sat
May 26Tue
  1. PAL39

    CENTAURO project improves TIAGo Pro's precision and safety in industrial handling

    The CENTAURO project, in which PAL Robotics participates, redesigned the 7-DOF robotic arm for the TIAGo Pro mobile manipulator, optimizing the workspace to reach angles and positions that were previously unreachable. The arm integrates series elastic actuators (SEA) that can absorb impacts and measure contact force, and shifts to torque and admittance control, enabling the robot to sense applied pressure. The WebGUI v3 ecosystem adds a drag-and-drop visual programming interface, allowing non-experts to configure complex tasks in scenarios such as polishing and logistics.

May 18Mon
  2. PKU Intelligent Robotics Open Laboratory20

    Zhang Wanruo from Peking University's Intelligent Robotics Lab has paper MXPose accepted by Pattern Recognition

    Zhang Wanruo, a 2021 direct-entry PhD student at Peking University's Open Laboratory of Intelligent Robotics, has had the paper "MXPose: Multiplex interactive learning for multi-view 3D Human Pose Estimation" officially accepted by Pattern Recognition (PR), a journal in the pattern recognition field. Her research direction is multi-view 3D human pose estimation.

May 15Fri
May 12Tue
Apr 27Mon
  1. PAL30

    StockBot Success Story: Redefining Inventory Accuracy in E-commerce Warehouses

    Fashion retailer Tendam, in partnership with PAL Robotics and ID Logistics, has deployed the StockBot autonomous RFID inventory robot at its logistics center in Illescas, Spain, and the solution is now fully operational. StockBot autonomously patrols daily, scans inventory with RFID, continuously supplements WMS data, and can identify items that are physically in stock but not registered in the system. After three years of gradual refinement from proof of concept, the partners say errors and incidents have decreased, order preparation is faster, and employees can shift to higher-value tasks.

Apr 22Wed
  1. New Scientist Robots76

    Sony AI table tennis robot Ace defeats professional players; research published in Nature

    Sony AI's autonomous table tennis robot Ace competed under the rules of Japan's professional table tennis league, defeating multiple elite and professional players; the research was published in Nature (DOI: 10.1038/s41586-026-10338-5).

    Why it matters: The article explains Ace's competitive record and three key technologies, and places it alongside the milestones of Deep Blue and AlphaGo, helping readers judge where physical sports competition currently stands.

Apr 20Mon
  1. PsiBot/灵初智能46

    Lingchu Intelligence releases self-developed complete robot ψ-SynRobot and starts mass production

    Lingchu Intelligence has officially released its self-developed complete robot ψ-SynRobot and simultaneously started large-scale mass production. The model targets logistics, retail, and industrial scenarios and uses an integrated "operation execution + data collection" architecture; each completed task simultaneously accumulates visual, tactile, and motion-trajectory data. It can run continuously 24/7 and suits scenarios such as warehouse sorting, retail guidance, and industrial assembly. The release gives Lingchu Intelligence a complete product matrix from an "embodied intelligence brain" to a complete-machine execution platform, covering forms such as wheeled robots and exoskeleton gloves.

  2. 普罗宇宙23

    Technology Empowers Intangible Cultural Heritage, Craftsmanship Passes Down Classics — ProUniverse Dabai Robot Appears at JD MALL

    The ProUniverse Dabai robot performed a dual-arm collaborative lacquer fan craft demonstration at the opening event of JD MALL (Wuhan Wangjiawan store), producing a customized lacquer fan within 1 minute. Its AI multimodal 3D vision and 2D industrial-grade positioning support sub-millimeter precision operation, and distributed learning enables rapid transfer of new scene skills. The event also introduced the "Craftsmanship Universe" plan, which, relying on the self-developed SDPAA parallelized data collection architecture and the InduThread-VLA large model, transforms traditional artisan skills into storable and replicable digital assets, with future expansion from crafts such as lacquer and sugar painting to precision assembly and high-end manufacturing fields.

Apr 16Thu
  1. OFweek Robotics50

    Youlu Robotics' AI130 Patrol Sweeping Robot Tops China's Outdoor Sweeping Robot Shipments in One Year

    Youlu Robotics, with its AI130 patrol sweeping robot, rose to the top of China's 2025 outdoor sweeping robot shipments within one year. Its core product is the Master2000 embodied brain with 254 TOPS of computing power, which uses a "one brain, multiple forms" model to adapt to different robot forms such as patrol sweeping and security patrols, and it has developed its own end-to-end general-purpose LPLM multimodal model. The company's "three gates" strategy progressively moves from public residential-community scenarios into unit doors and homes, feeding real-world scenario data back into the brain to achieve "deployment-free, out-of-the-box" delivery.

Apr 2Thu
Mar 17Tue
Mar 16Mon
  1. Dexterity66

    Dexterity speeds up Foresight visual perception 17x with NVIDIA, from 1508ms to 90ms

    Dexterity says that after NVIDIA acceleration, the visual perception pipeline of its world model Foresight cut per-cycle latency from 1508ms to 90ms, a 17x speedup.

    Why it matters: The original breaks down latency data and optimization methods for each stage of the pipeline, giving robotics developers a GPU acceleration approach they can adapt to their own perception systems.

Mar 11Wed
Feb 27Fri
Feb 17Tue
  1. RealMan/睿尔曼25

    Tech analysis: How many steps does RealMan's RealBOT wheeled folding robot take to "make dumpling filling" at the Spring Festival Gala?

    At the Beijing TV Spring Festival Gala, RealMan's RealBOT wheeled folding robot used remote teleoperation to open a refrigerator, take out soy sauce, pour soy sauce, and stir dumpling filling; it was the only company among the participating recording enterprises to complete on-site work via remote teleoperation. Its integrated joint module has an MTBF of 50,000 hours, opening the refrigerator door requires an instantaneous torque of about 14 kilograms, pouring soy sauce relies on visual guidance and high-frequency posture correction to achieve zero spills, and mixing the filling is controlled by a remote "isomorphic arm" that records data to feed back into training models for the remote operation network (GLN).

Feb 6Fri
  1. FieldAI51

    FieldAI and Boston Dynamics deploy autonomous robot inspections on construction sites

    FieldAI describes the autonomous deployment of its Field Foundation Models (FFMs) with Boston Dynamics' quadruped robot Spot on construction sites, requiring no GPS, cloud connection, prior maps, or pre-planned routes, and able to carry out inspection, mapping, and monitoring tasks around the clock. The article states that over the past two years this has covered construction sites in Asia, Europe, and North America, cutting customers' inspection and documentation time by more than 90%, with data able to flow into BIM and digital twin systems, and several pilots having expanded into fleet-level deployments.

Feb 3Tue
  1. 若创科技51

    Strutt launches Strutt ev¹ smart everyday vehicle at CES 2026

    Singapore robotics company Strutt officially launched the Strutt ev¹ smart everyday vehicle at CES 2026; the product previously won a CES 2026 Best of Innovation Award. The ev¹ features the evSense sensing system, which combines LiDAR, cameras, time-of-flight and ultrasonic sensors for 360-degree environmental perception, and supports Co-Pilot/Co-Pilot+ assist modes, Pathfinder autonomous navigation, and Waypoints natural-language command navigation based on a large language model.

Jan 27Tue
  1. Figure AI82

    Figure AI Releases Helix 02: A Full-Body Autonomous Control Model for Humanoid Robots

    Figure AI has released Helix 02, extending control from the upper body to the entire humanoid robot and enabling continuous, integrated autonomous tasks combining locomotion and manipulation. It adds a new System 0 whole-body controller, trained on over 1,000 hours of human motion data and simulation-based reinforcement learning, which outputs joint commands at 1 kHz via a 10M-parameter network, replacing roughly 109,504 lines of hand-written C++ code; System 1 connects head-mounted vision, palm cameras, fingertip touch sensing, and proprioception, outputting whole-body joint targets at 200 Hz.

    Why it matters: The original text provides details on Helix 02's architectural layers, training data, and demonstration tasks, allowing readers to understand the specific implementation approach for full-body autonomous control.

Jan 19Mon
Jan 7Wed
  1. Sharpa53

    Sharpa releases CraftNet, a hierarchical VTLA model for fine manipulation

    Sharpa has released CraftNet, an end-to-end hierarchical VTLA (vision-tactile-language-action) model aimed at last-millimeter contact control in fine manipulation. The model consists of three parts: System 2 (about 1 Hz, an open-source VLM responsible for reasoning and task decomposition), System 1 (about 10 Hz, responsible for motion planning and coarse actions), and System 0 (about 100 Hz, performing real-time fine-tuning of hand movements based on tactile feedback), with the three cooperating asynchronously.

Jan 6Tue
  1. Sharpa42

    Sharpa unveils North, its first full-body autonomous robot, at CES 2026

    At CES 2026, Sharpa unveiled North, its first full-body autonomous robot, and completed four fully autonomous live demonstrations. North has a 0.02-second reaction time in table tennis rallies, takes photos with about 2 mm precision, deals cards through multimodal reasoning, and completes a 30+ step paper pinwheel craft sequence, which the company calls one of the longest continuous autonomous operations publicly demonstrated to date.

  2. 可以科技31

    Meet LOONA DeskMate: The Desktop AI Companion for Work

    KEYi Technology introduced LOONA DeskMate at CES 2026 (January 6–9, Las Vegas), positioning it as a desktop embodied AI companion. It uses three signals—Intent, State, and Emotion—to achieve wake-word-free audiovisual multimodal perception, claims to eliminate 300–500 ms of silent latency, and connects to workflows such as Gmail, Slack, and Calendly through real-time clipboard and screen awareness.

Dec 19Fri
  1. Sharpa63

    Sharpa announces mass production of SharpaWave dexterous hand, to exhibit at CES 2026

    Sharpa announced that mass production of its flagship dexterous hand, SharpaWave, has begun, with early customers including global technology companies and research universities.

    Why it matters: The manufacturer announced that the dexterous hand has entered mass production and provided technical details such as degrees of freedom, tactile precision, and simulation platform compatibility, which readers can use to assess its research usability.

Dec 5Fri
  1. PKU Intelligent Robotics Open Laboratory41

    Professor Liu Hong's team wins Guangdong Provincial Natural Science Award First Prize for "Multi-Visual Feature Interactive Learning Mechanism"

    Professor Liu Hong's team's research achievement "Multi-Visual Feature Interactive Learning Mechanism" won the First Prize of the 2024 Guangdong Provincial Natural Science Award. The achievement reveals the mechanism by which prior knowledge guides the timing of robot multi-visual feature interaction, and has been applied to the industrial robot vision systems of ZTE Corporation and Skyworth Group production lines. Six representative works have been cited more than 5,000 times by over 630 research institutions.

Dec 3Wed
  1. Dobot/越疆机器人32

    Dobot Robotics appears at Geely Auto technology exhibition, embodied intelligent robots empower efficient flexible intelligent manufacturing in automotive

    Dobot Robotics recently exhibited at the forward-looking lean manufacturing technology exhibition hosted by Geely Automobile Research Institute, demonstrating on site the application in automotive manufacturing of a six-axis collaborative robot (equipped with the self-developed VX500 vision camera), the ATOM humanoid robot (the world's first full-size embodied intelligent robot with "dexterous manipulation + straight-knee walking"), and the Hexplorer six-legged robot dog.

Nov 26Wed
  1. Mentee Robotics · YouTube41

    Behind the Scenes: How MenteeBot Perceives and Plans Autonomous Logistics Tasks

    Mentee Robotics released a behind-the-scenes video showing how two MenteeBots combine perception, planning, and control in real time during a pick-and-place box logistics workflow; the left side of the frame shows the head-camera view, alongside the robot's "thoughts" and decision outputs. The clip comes from a complete 18-minute unedited autonomous run by two MenteeBots in the same setup.

  2. Mentee Robotics · X29

    Behind the scenes of MenteeBot box handling

    In response to questions from our previous tweet, we share a behind-the-scenes view of the same task. This video shows the view from MenteeBot's head camera in the top left and its "thinking" and decision outputs in the bottom left, directly presenting the perception and planning process as two MenteeBots carry out a box-handling task. Full uncut 18-minute video: https://www.youtube.com/watch?v=Ak0rswEQudE #humanoidrobots #PhysicalAI #AI

Nov 13Thu
  1. TechCrunch Robotics43

    Why a researcher is building robots that look and act like bats

    A team led by Nitin J. Sanket, a professor at Worcester Polytechnic Institute (WPI), has developed palm-sized flying robots that mimic bats' use of ultrasound, aimed at search-and-rescue missions. AI software filters noise from the ultrasound signals, letting the robots detect obstacles within a 2-meter radius. The team chose low-power ultrasonic sensors commonly used in automatic faucets, and to suppress interference from propeller noise, designed 3D-printed structures placed in front of the robot to modulate sound waves, similar to the function of a bat's nose, ears, and mouth tissues. The robot currently works; the next step is to increase its flight speed.

Nov 11Tue
  1. Keenon Robotics/擎朗智能47

    Keenon Robotics launches C55 cleaning robot for medium and large spaces

    Keenon Robotics unveiled the C55, a combined sweeping and washing robot for medium and large spaces, at the ISSA North America show, marking the product's first in-person debut in North America. The C55 targets indoor environments of over 3000㎡, offers a cleaning capacity of 2376㎡/hour, and features a 110L water tank, 21,000Pa suction, and fused perception combining 360° lidar and stereo vision. It supports automatic return-to-dock charging, wastewater discharge, and water refill, and its dual-battery system supports 24/7 continuous operation. The company also showed products including the C40, C30, W3, T10, and S300 at the same event.

Oct 30Thu
Oct 9Thu
  1. Figure AI67

    Figure AI releases third-generation humanoid robot Figure 03

    Figure AI has released its third-generation humanoid robot, Figure 03, redesigned from the ground up for the Helix VLA model, the home, and large-scale mass production, covering both software and hardware.

    Why it matters: The official announcement fully explains the new-generation humanoid robot's perception, touch sensing, home-safety design, and mass-production plans, so readers can understand its hardware changes and production capacity targets.

Sep 26Fri
  1. Keenon Robotics/擎朗智能47

    KEENON Robotics officially releases KOM 2.0, a self-developed VLA model for the service industry

    KEENON Robotics has officially released KOM 2.0 (KEENON Operator Model 2.0), which it calls the world's first VLA (vision-language-action) model for the service industry. The model uses a dual-system architecture with the K-Mind thinking system and the K-Act action system, is trained on hundreds of millions of real service interaction data points, and supports millisecond-level real-time decision-making.

Sep 19Fri
  1. LimX Dynamics66

    GVF-TAPE Algorithm from LimX Dynamics and Partners Accepted to CoRL 2025, Enabling Robots to Learn Manipulation from Video

    The paper "Generative Visual Foresight Meets Task-Agnostic Pose Estimation in Robotic Table-top Manipulation," a collaboration between SUSTech CLEAR Lab, LimX Dynamics, and the University of Hong Kong, has been accepted to CoRL 2025.

    Why it matters: This research presents a concrete technical approach and experimental figures for learning manipulation from human video, helping readers understand the feasibility of training robot manipulation with video data.

Sep 18Thu
  1. 线性资本72

    Bot Auto completes first fully driverless truck real-world road test

    Linear Capital portfolio company Bot Auto announced it has completed its first real-world road test in which the truck truly performed all operations on its own, with no one in the driver's seat, the back seat, or at a remote-control terminal, and no escort vehicle. During the test, the truck on a Texas highway handled scenarios including avoidance, detouring, merging, and responding to the appearance of a fire truck. The vehicle is equipped with 15 cameras, 8 lidars, and 3 millimeter-wave radars, with redundancy in perception, computing, actuation, and power supply; after the primary server is shut down it can switch to a backup server, and if the backup is also shut down, the vehicle control unit takes over to pull over and stop.

    Why it matters: The article sorts out the differences between Bot Auto's truly driverless road test and earlier modes with safety drivers, escort vehicles, and remote control, helping readers understand the differences in its technical approach.

  2. Figure AI72

    Figure launches Project Go-Big: Helix learns navigation zero-shot from human first-person video alone

    Figure announced two advances in Project Go-Big: a partnership with Brookfield to build a large-scale humanoid pretraining dataset, with Brookfield holding over 100,000 homes, 500 million square feet of commercial office space, and 160 million square feet of logistics space; and Helix, after training on 100% first-person human video, responding to natural-language commands such as "go to the fridge" without any robot demonstrations, navigating cluttered real home environments.

    Why it matters: The original gives the training method by which Helix learned navigation using only human video and the scale details of the Brookfield data partnership, offering a view of actual progress on the humanoid pretraining data approach.

Mar 21Fri
  1. Livox55

    Livox has delivered a cumulative 100,000 Mid-360 LiDAR units to Dreame, with a three-year target of one million units

    As of February 2025, Livox has delivered a cumulative 100,000 Mid-360 LiDAR units to Dreame. Under a three-year strategic partnership, the plan is to deliver more than one million units over the next three years. The sensors will be installed in Dreame's full lineup of robotic lawn mowers, covering more than 100 countries and regions, including high-end markets in Europe and North America.