NEURA Robotics announced that its NEURA Mobile Robots GmbH has acquired 100% of the shares of ADLATUS Robotics GmbH, an autonomous cleaning robotics company based in Ulm, Germany.
Why it matters: The original article provides details on the acquired company's existing equipment, navigation software, and plans to integrate with Neuraverse, which helps explain NEURA's cleaning business strategy.
World record! The Galbot robot completed more than 100 consecutive autonomous tennis rallies with zero errors! 🎾 Fully autonomous. 📷 Livestreamed. A new milestone for humanoid robot tennis! #AstraTennis #EmbodiedAI #HumanoidRobots #WHRG #Galbot
At the 2026 World Robot Conference, Wiz Robotics showcased two product lines: humanoid robots and intelligent industrial 3D cameras. The humanoid robot "Xiaowei" can respond to voice commands to grasp and deliver drinks, while the inspection robot is equipped with an AGV chassis and supports 360-degree rotation; the intelligent dexterous hand has 16 degrees of freedom, fingertip pressure of 12N, and a maximum lifting load of 15kg. The industrial 3D camera matrix covers interpupillary distances from 90mm to 620mm, including the "Starlight Eye" welding series and the "Smart Light Eye" grasping series, and on-site demonstrations included teaching-free corrugated board welding and 3D point cloud dynamic imaging.
From AlphaGo to AstraTennis. In 2016, AlphaGo showed that AI could master the digital world. Ten years later, AI has entered the physical world. Today, at the World Humanoid Robot Games, the Galbot robot completed the world's first live autonomous humanoid robot tennis match, competing alongside human athletes. 🎾🤖 The Galbot robot performs vision, perception, movement, and competition in real time! This is the AstraTennis moment. #WorldHumanoidRobotGames #WHRG #Galbot #Humanoid #Robot
At WRC 2026, PaXini released its fourth-generation GEN4 multi-dimensional tactile sensor matrix, featuring the self-developed fourth-generation native true 6D tactile sensing chip FUSE, adding the Superior flagship sensing unit, Neuron full-body electronic skin, and Gripper perception module, and offering 19 sensor options in different sizes.
On August 19, Professor Zhao Mingguo's team in Tsinghua University's Department of Automation published the paper "Learning Vision-Driven Reactive Soccer Skills for Humanoid Robots" in Science Robotics, selected for the August 2026 humanoid robotics special issue; the robot platform used was a humanoid robot independently developed by Booster.
Why it matters: The paper breaks down three failure points of the traditional perception-planning-control pipeline and presents a reproducible sim-to-real transfer method.
How can a robot clear a cluttered tabletop? 🤖 Our Physical AI system lets the robot recognize each item, determine where it belongs, and carry out a long sequence of continuous actions, including coordinated two-arm handovers. Thread 👇
PaXini AI released the five-finger data collection glove PXCap Pro on August 14, integrating tactile, visual, and joint posture modalities and focused on high-consistency physical ground-truth data collection. It offers 0.1N tactile sensitivity, a 165° ultra-wide-angle wrist camera, built-in ±0.15° angle encoders, and weighs about 325g, achieving millisecond-level hardware synchronization across modalities. The product will be showcased at Booth B203 in Hall B of the 2026 World Robot Conference (WRC), which opens on August 19.
Daimon Robotics released its tactile-anchored world model Daimon-TWM on August 10, and completed a strategic financing round of several hundred million yuan the next day, its second hundred-million-yuan-level financing within two months.
Weijing Intelligent presented the Starlight Eye ProV2 weld-seam dual-measurement camera at the 2026 Fifth GG Intelligent Welding Robot Summit held in Shanghai on August 12. The product integrates six functions in one: static area scanning, dynamic line scanning, precision RGB imaging, weld-seam tracking, molten-pool monitoring, and dynamic weld-seam imaging. Without relying on 3D drawings, it can autonomously acquire 3D weld-seam data before welding, perform real-time tracking and molten-pool monitoring during welding, and help build a PQC quality closed loop. It also builds a welding ecosystem through an open SDK and algorithm interfaces. Weijing Intelligent also received the "2026 China Intelligent Welding Industry Outstanding Product Award" at the summit.
Mila-Quebec AI Institute and a Polytechnique Montréal team have released Milo, an open-source robot guide dog that they say can cut the roughly $50,000 cost of a guide dog to about $2,000. The hardware base is a Unitree Go2 Air/Pro plus an NVIDIA Jetson Orin Nano; a magnetic-encoder telescoping handle is used to inversely solve the user's 3D position, a bird's-eye-view map is built first and then a navigation policy is run, and the policy is trained in about 10 minutes in GPU-accelerated simulation and deployed zero-shot on the real robot.
Why it matters: The original article fully breaks down Milo's perception, policy training, and real-world testing details, so readers can reuse its engineering trade-offs in their own navigation projects.
In its home breakfast service scenario, UniX AI's Panther humanoid robot autonomously plans and completes the full workflow, from preparing ingredients and heating food to making drinks and serving the meal at the table. UniCortex handles long-horizon task planning, UniFlex supports rapid learning of new household skills from a few demonstrations, and UniTouch combines visual and tactile perception to improve manipulation reliability for food, kitchen tools, and everyday objects. More information at www.unix-group.ai.
PaXini has released PX-FOOTRIX, a multidimensional plantar tactile sensor. Built on 6D Hall-array tactile sensing technology, it enables full-sole three-dimensional array force sensing and six-axis force/torque perception, helping bipedal robots perceive ground characteristics and optimize gait. The product offers IP67/68 custom protection and 1000% impact overload capacity, supports SPI, UART, RS485, USB, I2C, CAN FD and other communication protocols, is plug-and-play, and will be showcased at WRC 2026, Hall B, Booth B203.
The Hongxing Technology Xiaomei M9 humanoid service robot focuses on continuous service off-grid and targets elderly care, medical, government, and commercial service scenarios. Its perception, modeling, navigation, and decision-making are 100% completed on local edge computing, enabling it to keep working for more than 6 hours under disconnected or weak network conditions, with zero data leakage. Its millimeter-wave radar fall prediction has an accuracy rate of 99% and a false alarm rate below 1%, and its robotic arm completes the entire process from perception to intervention in just 10 seconds.
On August 6, Chengxin Zhilian held a centralized delivery of nearly 100 industrial inspection quadruped robots at the Hubei Humanoid Robot Innovation Center in Wuhan Optics Valley. The first batch of 30 Xun 1 series units is priced at over 100,000 yuan each and achieves millimeter-level positioning accuracy. The simultaneously launched Xun 3 is equipped with a 32-line LiDAR, can climb 25 steps and ascend a 35-degree slope, and has increased battery life from 1 hour to 4 hours, enabling continuous walking of about 20 kilometers. The company develops its own MEMS inertial navigation modules, with annual capacity of 2.6 million units, overall unit monthly capacity of over 100, and orders booked through the fourth quarter of this year.
Through the Draper software update, Skydio is delivering free camera performance improvements to in-service X10 drones via OTA, with no hardware replacement required. The update includes low-light tuning and autofocus improvements for the telephoto camera, tap-to-focus in Remote Flight Deck (hold the F key and click a target on screen to adjust focus and exposure), a dedicated low-light camera mode, and, during Shadow tracking, target-anchored linked focus, exposure, and stabilization.
In an interview, Boston Dynamics mechanical engineer Taylor Frey-Baker discusses the design of the electric Atlas's head and how she entered the field. She is responsible for the compute core inside the head, which houses four perception modules with stereo vision; the head's modular design allows the entire unit to be swapped out and the compute core upgraded like a game cartridge. The hydraulic Atlas did not need a head because it performed movement demonstrations, whereas the electric Atlas relies on head cameras for a wider field of view and uses lights and sound to signal its status to people nearby.
BeingBeyond has released Being-H0.8, which it calls the world's first implicit tactile world action model, unifying vision, touch, action, and future state changes into a single latent space, with pretraining on more than 500,000 hours of first-person human video.
Why it matters: The model brings touch into large-scale pretraining along with a self-developed pseudo-labeling system, letting readers understand the path by which tactile data goes from a lab signal to a scalable pretraining resource.
Enabot has released a comparison guide for the EBO Max and EBO X: the EBO Max uses embodied AI multimodal intelligence, with environmental perception, long-term adaptive memory, and two-way video calling, but requires manual guidance for mapping; the EBO X is powered by ChatGPT-driven conversation, supports Alexa integration and V-SLAM automatic mapping, and offers fall detection, cry detection, and help-call alerts (effective detection range 10–15 feet). The two differ in privacy approach: one focuses on software management, the other on a hardware physical button.
Using the AgileX UMR industry-grade universal modular mobile robot chassis and the AgileX NAVIS 3D 3D laser navigation platform as examples, AgileX Robotics argues that navigation is evolving from project-level algorithm development into reusable infrastructure. The article holds that navigation time is mostly spent on engineering integration such as sensor synchronization and coordinate frame management rather than on the SLAM algorithm itself; in dynamic, open, and three-dimensional spaces, 2D environmental information can no longer describe the real environment, so 3D laser navigation has therefore become a new foundational capability.
Yaskawa Electric has developed the "Agentic Robot System," which links the AI robot MOTOMAN NEXT with Google DeepMind's generative AI model Gemini Robotics ER 1.6. The robot can assess conditions on site on its own, assemble and execute work steps, and complete tasks autonomously from instructions such as "sort these parts," without prior teaching or detailed programming.
Tashan Technology recently announced the completion of a Series B financing round worth hundreds of millions of yuan, with investors including Lavender Hill Capital Partners (LHCP).
Why it matters: Operating data of tens of thousands of units shipped per month and a share exceeding 80% lets readers gauge the actual progress of tactile sensing toward scaled deployment.
Huiwen has launched an industrial robot product matrix covering three product lines: the iBen X series, the iBen P series AMR, and rail transit robots. The iBen X series supports loads from 100 kg to 1500 kg, and the iBen P series covers 1000 kg to 3000 kg; the iBen P1T model uses laser SLAM navigation to achieve ±5 mm positioning accuracy. The rail transit product line covers transport, inspection, and cleaning; the undercarriage train inspection robot weighs less than 30 kg, has an inspection speed of 10 m/s, and outputs detection results in 30 seconds.
1X Technologies announced a 25-DOF tendon-driven dexterous hand for the NEO humanoid robot platform, which will ship with every NEO unit.
Why it matters: The original article provides key specifications and production progress for the 25-DOF tendon-driven hand, which readers can use to assess its actual impact on humanoid manipulation capabilities.
PKU EPIC Lab proposes WAM-TTT, a test-time training framework that steers world-action models from raw human videos without robot actions, human-side annotations, or task-specific fine-tuning. The method absorbs human videos into a lightweight adaptive memory inside a frozen WAM through self-supervised video prediction, and uses a meta-training stage with paired human-robot data and a key-value memory reconstruction objective to align human demonstrations with robot behaviors. Experiments show it consistently outperforms in-context human-video conditioning baselines across diverse manipulation tasks and generalization settings.
Why it matters: The original work proposes a method for adapting a frozen world-action model at test time using only unlabeled human videos, without robot demonstrations or fine-tuning, offering methodological reference value for researchers who want to reuse it.
LingBot-Depth 2.0 has been released, a next-generation spatial perception model for robots that focuses on perception of glass, mirrors, and transparent objects. It was trained on 150M-scale data (50 times that of v1.0) and took first place in 12 of 16 depth completion benchmarks; on the hardest indoor scenes, depth error RMSE dropped from 0.132 to 0.062.
Why it matters: The original text provides benchmark rankings, the change in depth error on the hardest indoor scenes, and an open-weights link, helping readers judge the improvement over the previous generation and how reusable it is.
Ant Lingbo Technology has launched LingBot-Map, which uses Geometric Context Attention (GCA), an end-to-end learned streaming state, to solve the memory problem in streaming 3D reconstruction.
AgileX Robotics' self-developed NAVIS 3D laser navigation platform combines a 16-line LiDAR with visual sensors, can build maps at an operating speed of 1.5 m/s, achieves static repeat positioning accuracy of ±5 cm, and supports 3D map construction for scenarios at the 2 million square meter scale. The platform works in coordination with AgileX's UMR all-scenario universal four-wheel-drive four-wheel-steering mobile robot chassis, and an intelligent under-track inspection solution for rail transit with a dual-arm lifting mechanism has already been launched, integrating the PiPER dual robotic arm system and dual 32-line LiDARs to achieve a fully automated closed loop for the entire under-track inspection process.
Boston Dynamics' Spot robots are deployed at the Dallas International Broadcast Center and New York/New Jersey stadiums, carrying out perimeter patrols to support asset protection and risk detection for the FIFA 2026 World Cup. Spot 5.1, released this year, supports security missions: when it detects a person, it pauses, activates its AV lights, captures PTZ, panoramic, and thermal images, and triggers an alarm. The robot is positioned as an assistant to security personnel, with patrol findings ultimately reviewed by staff.
The Chinese University of Hong Kong, Shenzhen and other institutions propose GeoPredict, which explicitly predicts end-effector and joint-level future trajectories before the VLA generates actions, and uses 3D Gaussian modeling for the geometry of the future workspace.
Why it matters: GeoPredict incorporates future trajectory and 3D geometry prediction into VLA training, giving readers a spatial reasoning approach that does not increase inference overhead.
STIER uses AI flexible grinding technology to address the pain points of easy scratching and insufficient consistency in weld seam grinding for battery box enclosures. Its solution uses PolishX flexible force control (accuracy ±0.1N) and millimeter-level visual positioning perception, combined with a process grinding knowledge base and the NextBrain™ AI cloud system to make grinding decisions, and has self-learning and adaptive capabilities. It also provides full-process support including customized laser scanning solutions, real-time data monitoring, and grinding head wear warnings.
STIER has launched an AI flexible grinding robot workstation that builds a "perception-analysis-decision-execution" physical AI closed loop, targeting the grinding of multi-curved-surface parts in high-end manufacturing such as aerospace and precision molds. The solution uses 3D vision inspection and positioning, a PolishX flexible force-control system belt grinder, and the self-developed NextBrain™ Mushroom Cloud system. In a wet grinding case involving silicon carbide composite workpieces, it achieved fully autonomous operation across the entire process, with contact force control accuracy of ±0.5% FS, per-piece processing time shortened by more than 22%, and product changeover completed within 10 minutes.
JAXA researchers reported in Science Robotics that the palm-sized spherical lunar rover LEV-2 (nicknamed SORA-Q) operated autonomously for more than 100 minutes after landing near Shioli Crater during the January 2024 SLIM mission, traveled an estimated 24 meters, and returned multiple images.
Deep Robotics' wheeled-legged robot Lynx M20, equipped with a smart utility tunnel solution, has been deployed for inspection in nearly 26 kilometers of underground utility tunnels at Chengdu Future Science and Technology City. The robot combines wheeled and legged locomotion, can pass through passages as narrow as 50 centimeters, and supports autonomous charging and automatic return for 7×24-hour inspection. It integrates infrared temperature measurement, gas detection, high-definition cameras, acoustic collection, and other sensor modules, and works with Junyi Digital to carry a digital twin operations and maintenance system, enabling three-tier red, orange, and yellow alerts and a closed digital inspection loop.
The Strutt ev¹ firmware 1.2.3.11 update brings a new customizable UI, enhanced evSense, and a safety confirmation mechanism. The new UI supports a dynamic home screen that changes color by driving mode, a pull-down dashboard, and 9 customizable shortcuts; uphill power on slopes increases by up to 25%, and evSense is ready within 5 seconds of waking. Enabling features such as Level S and Hands-Free Auto-Pilot requires completing safety protocol confirmation in the app; Level S is not supported in Singapore and Japan.
Enchanted Tools demonstrates the Mirokaï humanoid robot turning a spoken request into an autonomous process of grasping, carrying, and delivering an object to a designated area. The demo covers natural language understanding, autonomous grasp planning, drop-off area detection, and precision manipulation; the robot combines perception, planning, and manipulation to complete a real-world task. The source does not provide quantitative metrics such as success rate or payload. The post is dated June 16th, 2026.
Investor Shi Ma speaks with Stanford PhD student Huang Wenlong (advisor Li Feifei) about his PointWorld work and 3D world models. PointWorld uses 3D point flows to unify state and action representations, mapping different robots' joint actions into point cloud changes via URDF to address inconsistent actions across embodiments; training used the DROID and BEHAVIOR datasets, with annotations relying on FoundationStereo, VGGT, and CoTracker3.
The Fuwei Intelligence Fuzhi 1 and Fuzhi 2 are mass-produced industrial humanoid robots, each with a dual-arm rated payload of 8 kg. Fuzhi 1 is equipped with FW75 precision robotic arms and dual RGB-D depth cameras, with accuracy of ±0.02 mm, and focuses on fine assembly; Fuzhi 2 has FW65 heavy-duty robotic arms, three industrial cameras, and a 2.2-meter lifting column, and focuses on heavy-load transfer. Both robots share Fuwei's self-developed AI large model and digital twin simulation platform, and are suited to industrial scenarios including automotive, electronics, logistics, healthcare, new energy, and CNC machining.
Enchanted Tools' service robot Mirokaï demonstrated semantic location mapping, natural voice commands, and autonomous navigation: after associating places with meaningful names, the robot can understand requests such as "go to the café" or "go to reception" and autonomously navigate to the destination without coordinates or programming, described as a human-robot interface achieved through conversational interaction. The video ends with a teaser for a milestone on June 16, 2026, and the message "WE'RE READY," while inviting viewers to join its private Discord community for behind-the-scenes content and early access.
At the inaugural EgoCross Challenge of the CVPR 2026 EgoVis Workshop, Nuoyin Intelligence took first place globally in both the Source-Limited and Open-Source tracks, becoming the only team to win championships in both tracks.