By 2035, 900 billion AI agents will run robot cafés, autonomous farms, and the cars that drive you home. But how do billions of independent machines agree on what happens next without a central brain? They speak Tashi.

Tech blogger 神奇阿宇 from Hubei, China, built a working hoverboard. Electric, with 10 brushless motors pushing over 30kg of thrust each—enough to lift 4 people—and peak current hitting 2,000 amps. 2 STEPX Neo phones as the board's brain. One inside the board reads posture data in real time; the other on his wrist feeds AI voice prompts through his earbuds—"Weight's shifting left!" "Bend your right knee!" After 10+ days of brutal training and AI-assisted correction, he achieved man-machine sync.

15 Robotics Research Voices on X @chris_j_paxton = Honest takes on every humanoid demo, from AI at Agility. @KarlPertsch = Robot datasets and open VLAs, from DROID to OpenVLA. @kevin_zakka = The sim tooling behind MuJoCo Menagerie and mjlab. @chichengcc = Diffusion Policy and UMI, from the person who built them. @hausman_k = Robot foundation models, from RT-2 to Physical Intelligence. @ankurhandos = Dexterous robot hands, straight from NVIDIA research. @du_yilun = Video generation models used as robot planners. @AjayMandlekar = A few human demos turned into thousands in simulation. @0x_meden = Robotics jobs, free courses and open-source robots, broken down. @xuxin_cheng = Humanoid teleoperation, from the creator of OpenTeleVision. @peteflorence = Scaling robot foundation models at Generalist. @breadli428 = How legged robots and humanoids learn to move. @ericjang11 = Long essays on where robot learning is heading. @IlirAliu_ = The robotics industry for founders, investors and operators. @TheHumanoidHub = Every humanoid launch, demo and funding round in one feed.

Working with robots has always been harder than it should be. The intelligence was hard to access. The hardware was hard to afford. Today, both change. @makermodsai Maker Arm + Metal Arm, powered by @dreamlabs_ai 's Dream-1. Give your robot a task in plain language, right from your browser. Try it: playground.makermods.ai

Inspired by feather stars, the NC State researchers report a soft underwater swimming robot capable of achieving three distinct swimming modes—jellyfish-like vertical swimming, fish-like directional steering, and rotor-like 3D reorientation—using only two pneumatic inputs. By embedding mechanical intelligence into its elastic structure, the robot seamlessly transitions among these modes for agile and versatile underwater maneuvering. Particle image velocimetry (PIV) and computational fluid dynamics (CFD) reveal the distinct hydrodynamic mechanisms underlying each swimming mode. The multifunctional swimmer can also be equipped with magnets, hooks, and miniature cameras for underwater waste collection, cooperative manipulation, environmental monitoring, and exploration. This work offers a promising platform for future adaptive and autonomous soft robots capable of navigating and performing tasks in complex and unstructured underwater environments. Credit: Jie Yin, Haitao Qing, Matt Shipman #engineering #technology

Collaborative proprioception — where robots know where other robots and devices are without relying on their camera — is going to enable sci-fi levels of autonomous coordination. Only on the real world web.

Robots will need data. Machines will need identities. And both will need an economy On Monday, @openroboto joins us to talk about how that economy gets built, and the part we’ll play in it together 📅 Oct 19 · 11:00 CEST xitter.cf/i/spaces/1yKAPwVazayxb

Deploy, deploy, deploy... Korea Clears First 20 Humanoid Robots for Real Factory Lines Alongside Workers South Korea’s Ministry of Trade, Industry and Energy has approved a regulatory sandbox exemption for Robros to deploy 20 AI bipedal humanoids on live production lines without traditional safety fences. The four sites-> >Nepes (semiconductor): Clean contaminants from circular components in back-end packaging. >Sungwoo Hitech (auto parts): Place body components into designated positions. >HD Hyundai Samho (shipbuilding): Stack scaffolding brackets used on ship hulls. >Delivery Lab (logistics): Move refrigerated packaging units. The robots must move, perceive their surroundings, and operate near people rather than stay fixed at one station. Test data will feed into revisions of existing safety standards. They are Robros’ IGRIS-C (154 cm, ~56 kg)--compact for factory aisles, ~31 body DoF, tendon-driven hands, ~6 kg dual-arm payload--with RL locomotion, imitation learning, and full-body teleoperation. IGRIS-C has already done earlier trials in logistics（as in the video shared earlier）; this is a larger-scale sandbox test on real factory floors. And,this is also Korea’s first broad clearance for humanoids working on the same lines as people, focused on mature industries with existing production lines.