Booster Robotics has launched the Booster Championship Night Simulation Season 1 3v3 soccer tournament and is recruiting participants. For product inquiries, contact 400-865-2028 or sales@booster.tech.
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.
Open source is more than code. 🤖 • Last week, we welcomed the French ROS community to our headquarters for the @ROSConFr social event.
It was a lovely evening of continued conversation after two days of talks and workshops.
See you next year! 💜
#ROSCon #robotics
EletrolarShow has come to a successful close. 🇧🇷🎬
From live demos to late-night conversations about emotion-driven robots, thank you to everyone who made this unforgettable week so great.
We return with inspiration from Brazil. More to share soon.
#Zeroth #BrazilExpo #ActiveEmpathy #ThankYou #Recap
AgileX Robotics released a tutorial explaining how to build a complete data collection solution for mobile manipulators based on LeRobot, noting that most public datasets in the LeRobot ecosystem come from fixed desktop robotic arms and lack data on mobile capabilities.
The Yuyao Robotics Research Center has announced Hua Zhehao as the proposed hire for its non-establishment recruitment, for the position of AI application R&D engineer, with an interview score of 88 and passing both the medical examination and the review. The public notice period is 3 days, and objections may be raised via the designated email address.
Bear thought it could bully the rabbit? Not with #Eiliko's dad around. After bootcamp, it finally came down to 🐻 versus 🐇. The secret move is — a smarter escape! A tactical retreat is Dad's top-tier lesson. Happy Father's Day to all the fathers who have our backs at the critical moments and run alongside us!
Cheer alongside Dad.
Root for your favorite team.
And companionship that makes every moment meaningful.
#ollobot #cyberpet #footballsunday #happyfathersday
Happy Dragon Boat Festival 🚣♂️🍃 Drums are rolling and the streets are filled with the smell of zongzi! Inspired by the dragon boat races, our little Eilikos built their own dragon boat, loaded with a boatful of special blessings for you. #Eilik, #Eiliko and the Energize Lab team wish you health and peace. Happy Dragon Boat Festival! 💚
Nuoyin Intelligence has teased that it will unveil an embodied robot at WAIC 2026 on July 17, 2026, calling it the world's first scale-disruptive embodied robot, focused on balancing extreme stowability with full-area operation in indoor scenarios. The original text provides only teaser information and does not disclose a specific model, specifications, or performance data.
Commercial space in-orbit operation and management service company Aerospace Satellite recently announced the completion of its Series D, D+ and D++ financing rounds, with cumulative funding of nearly RMB 2 billion. Participants include Puhua Capital, Sequoia China, Qianhai Ark, Unicom Strategic Emerging, Zijin Mining, China Insurance Investment, SAIC Motor, and Wuliangye, among others. After completion of the D+ round, the company joined the ranks of unicorns valued at RMB 10 billion.
On June 10, Harbin Institute of Technology and the Fuzhou New Area Management Committee signed a special cooperation agreement on constellation construction, and Lan Wen's delegation visited the State Key Laboratory of Robotics and Systems and other units. The two sides will jointly build a commercial space satellite constellation for marine applications and an innovation cooperation platform in the satellite field, promoting the commercialization of satellite-related scientific and technological achievements in Fuzhou New Area.
Diffusion (or flow) makes for excellent policies, but training them with RL is notoriously hard: BPTT is unstable, and RL over diffusion blows up the horizon.
In our new paper, we show how to optimize a flow matching actor with a "weird little trick"—using the identity matrix to "approximate" the Jacobian of the flow denoising process. 👇
ETH Zurich Robotic Systems Lab congratulates Andreea Tulbure on completing her doctoral defense; her thesis focuses on how legged manipulation robots can perform object handovers with humans in an intuitive, task-oriented manner. The original text does not provide specific methods or experimental metrics.
@LuLu_Mall has set a new benchmark for every #mall in the country.
#Odigo brings a real upgrade to the mall experience, keeping customers engaged, informed, and willing to come back.
Contact us to make this happen for you too.
#LuluMall #MallInnovation #RetailTechnology #SmartRetail
Scaling reinforcement learning to long horizons remains a major challenge.
Long-horizon Q-learning (LQL) prevents the accumulation of bootstrapping error by constraining differences in the value function over long horizons.
Compared with one-step TD and n-step returns, LQL shows significant improvement!
Paper: https://arxiv.org/abs/2605.05812
Today (June 3), I will be speaking at the CVPR "Test-Time Scaling for Computer Vision" workshop (1:30 pm PT) on how we can use test-time compute to improve the generalization of robot policies, Room 506.
Also, I am speaking right now (in 5 minutes) at the "Foundation Model Deployment" workshop in Ballroom 2!
Noble Machines CEO and co-founder Wei Ding will speak on Wednesday at 11:45 AM CT on the Humanoid Pavilion Stage at Automate 2026, on industrial autonomy driven by scalable robot learning. He noted that traditional automation relies on extensive programming, fixed workflows, and long deployment cycles, making it hard to handle dynamic and constantly changing industrial and manufacturing scenarios. The company is also offering a booking link for 1:1 meetings in Chicago.
Attending @CVPR and want to explore the sparks between vision, world models, and robotics?
We will host a small reception in Denver on June 6, and @DynaRobotics' co-founders and research team will all be there.
Come have a beer, hang out, and debate whether vision models can become systems that understand and act in the physical world.
June 6, Denver. Limited spots—yes, we are intentionally keeping the numbers small.
RSVP: https://luma.com/rcz2dzht
The 2026 graduating class of Peking University's Intelligent Robotics Open Laboratory completed their thesis defenses on May 17. Three doctoral students, Zhang Wanruo, Liu Jinfu, and Wang Wenshuai, were recommended for conferral of Doctor of Science degrees, and five master's students, Jiang Peifeng, Chen Siyu, Tang Tao, Xu Jing, and Xu Xinhua, were recommended for conferral of Master of Science degrees. Zhang Wanruo's doctoral thesis was recommended as an outstanding doctoral thesis, and Tang Tao's master's thesis was recommended as an outstanding master's thesis; their advisor is Professor Liu Hong.
A UC San Diego team created microrobot swarms made of living algae and nanoparticles that can switch between gathering and dispersing using light color: blue light makes them assemble into specified shapes according to a mask, while red light makes them disperse. In simulated wound tests, AI analyzed the wound and laser-printed a matching mask; under blue light the robots assembled onto medical tape in the shape of the wound, and after red light irradiation, over 90% were released to the target area in under two minutes. The authors note that the system currently relies on light, is limited to surface applications, and still requires loading the nanoparticles with drugs and testing on real wounds and living tissue.
Genesis AI's robot performs lab pipetting, liquid transfer, and mid-air manipulation of delicate instruments in autonomous mode at 1x speed, using its robotic foundation model GENE-26.5. The demo targets fine motor control, a highly demanding test scenario, and showcases the model's manipulation capabilities in high-precision laboratory work.
America's largest autonomous lab, from general AI to specialized intelligence 🦾 and more | Neo Highlights
Check out our summary of April updates and announcements from Neo portfolio companies and community members 💙
https://open.substack.com/pub/neo/p/americas-largest-autonomous-lab-from?r=6xf9ka&utm_campaign=post&utm_medium=web
Researchers at EPFL published a method in Science Robotics that lets multiple robots with different designs complete the same task from a single human demonstration. The method does not rely on AI; instead, it analyzes the physical properties of three-revolute-joint robotic arms, marks safe regions within each robot's range of motion, and classifies three-joint robots into six categories based on shared physical constraints, writing those constraints into each robot's program. The authors call this kinematic intelligence.
I will give a talk tomorrow (Monday) at ICLR on long-term memory for long-term autonomy.
9:25 AM @ MemAgents workshop
https://sites.google.com/view/memagent-iclr26/home