Affgan

[Affgan @treck658]

@GavinPartridge Sent. Thanks Gavin. Great channel 👍

[谢小总 @xie_Dl]

这工作连机器人都干得有情绪了…

[The Humanoid Hub @TheHumanoidHub]

Pen spinning is a classic test of human finger dexterity. WUJI just trained a robot hand to do it.

Their sim-to-real policy, trained entirely in simulation, spins a pen on a physical WUJI Hand 2. At IROS 2026 they also showed glove-based teleoperation, a human driving the hand's motions live.

Two paths to dexterity: learn it in sim, or lend it from a human.

[Dominic Lee 李梓敬 @dominictsz]

Robots are now sorting parcels in Chinese logistics centers!

They work fast and keep things moving smoothly!

[Mick Ryan, AM @WarintheFuture]

A Ukrainian large drone airlifts a four-legged combat robot. This mothership approach to delivery of UGVs into combat has come of age in 2026. Are Western military institutions learning?

[Affgan @treck658]

@ianfromphilly You must be doing something wrong, Fable and Astra are incredible on excel

[Yura_5800 @Yura_5800]

The startup behind this pink robot has raised $200 million in investment in the past 18 months — and no trucker knows about it yet.

The money didn't go to marketing or an office in downtown San Francisco.

It went to one specific task: teaching the robot's arms to lift heavy, uneven, slippery objects — something that was previously considered "impossible for machines."

In this video, the robot demonstrates exactly what made investors buy into the idea with 9 zeros.

The question is simple: how many more rounds of funding will it take before robots like this become cheaper than a trucker's salary?

[Lukas Ziegler @lukas_m_ziegler]

Balloon-powered robots! 🎈

Surely one of the coolest and craziest robot project I've seen this year.

Buoyant Choreographies by RoMeLa is a fun art project shown at ICRA in Atlanta last year.

It uses special robots made from helium balloons and moving legs that float and walk around.

People can play with the robots using touch, game controllers, or even by blowing air.

The robots react by changing colors, moving differently, and showing emotions.
@DennisHongRobot, that's so cool, what's next?! 🤯

Here's the paper behind it: https://t.co/jDQESthPvS

~~

♻️ Join the weekly robotics newsletter, and never miss any news → https://t.co/GoA3ZuwoPB

[Sentinel Robotics @sentinelcorp_us]

I don't think people understand how monumental this kind of thing is. Harvest has always been the hardest part of agricultural automation and leafy greens are the hardest of the hard.

Amazing stuff happening out at Reservoir!

[Techniahqrobot | humanoid robots @techniahqrobot]

This might be the cleanest performance of the entire World Humanoid Robot Games.

Watch this.

[Kyle Chan @kyleichan]

This is the most impressive to me, more than the 100m world record.

This requires autonomous real-time planning and action in response to a fast-moving target and dynamic environment.

Galbot is one of China’s top humanoid robot startups.

[Amani @SoyAmanii]

Esto es jodidamente peligroso.

Apple pasó 7 años y gastó $3,500 por unidad desarrollando el Vision Pro. Este tipo hizo algo mejor con una simple webcam desde su habitación.

Y no, esto no es un render. Está funcionando en tiempo real en su laptop ahora mismo, con una webcam apuntando directamente a sus manos.

La cámara rastrea 21 puntos de cada una de sus manos en tiempo real: cada punta de los dedos, cada nudillo y cada articulación, todo mapeado sobre un esqueleto 3D a 60 fotogramas por segundo.

Una pieza de seda digital está vinculada a ese esqueleto. Cuando abre la palma, la tela cae sobre sus dedos. Cuando cierra la mano, se arruga dentro de su puño. Cuando inclina la muñeca, la tela se desliza y se dobla utilizando física real, mientras la luz rebota sobre ella como si fuera un material de verdad.

Puede agarrarla, estirarla, lanzarla y atraparla, todo usando únicamente sus manos y sin llevar ningún dispositivo encima.

Sin visor.
Sin guantes.
Sin una computadora de $3,500 pegada a la cara.
Sin un ladrillo de 600 gramos sobre la cabeza.
Sin una batería que dura 2 horas.
Sin Apple ID.
Sin App Store.

Solo una webcam, TouchDesigner y un chico que vio la presentación del Vision Pro y pensó:

“Yo puedo hacer eso gratis.”

Apple tiene 3,000 ingenieros, $200 mil millones en efectivo y 7 años de desarrollo. Construyeron una computadora para la cara que la mayoría de las personas probablemente nunca tendrá.

Este tipo tenía una laptop, una webcam y un fin de semana.

Y construyó un futuro que la mayoría de las personas realmente podría usar.

Eso es lo que pasa cuando dejas de esperar permiso y simplemente empiezas a construir.

[el.cine @EHuanglu]

wwow.. AI just turned red alert 2 into 22mins epic film

[Lukas Ziegler @lukas_m_ziegler]

The robot that pinches flowers! 🌷

A robot pinches chrysanthemums slower than a skilled worker on a good day.

But... that's missing the point entirely. Because it's the repeatability + quality of pinch that is crucial in this case.

The same pinch, thousands of times. Plant #50,000. Identical.

A tired worker on hour seven cuts differently than hour one. Muscle fatigue changes precision.

tHe rOboT dON't FaTiGue!!!!!!!!!

Quality control becomes predictable, because you know that consistency will make it easier :D

You hear this every week: "The robot is slower than my best operator." Sure. On Tuesday when they had good sleep and coffee.

~~

♻ Join the weekly robotics newsletter, and never miss any news → https://t.co/GoA3ZuwoPB

[Aditya Bhatt 🤖🛹 ✈️ CoRL @aditya_bhatt]

A first glimpse of the wildest robotics project of my PhD.

🛹 The Blind Skateboarder 🛹

No vision. No markers.

Paper, code, weights — coming soon. And we have more. 😉

[JT @trev3d]

Impromptu testing in the park. Hoping to have Lasertag 3.0 out soon™

[JT @trev3d]

Zombies

[Simon Kalouche @simonkalouche]

Figure robots officially unemployed.

[Ilir Aliu @IlirAliu_]

The most interesting part of this robot is not the arm.

It is the gripper.

Instead of pushing the whole robot faster, this setup adds speed at the end effector.
The gripper itself becomes a moving axis that can accelerate independently of the arm.

That changes the physics of the system.

• Faster cycle times without stressing the robot structure
• Less inertia to fight against
• Motion where it actually matters, at the tool center point

It is a reminder that many performance limits in robotics are not solved by bigger motors, but by smarter mechanics.

——-

Weekly robotics and AI insights.
Subscribe free: https://t.co/9Nm01QUcw3

[Fabrizio Bustamante Escudero @Fabriziobustama]

This is where humanoid robots start becoming economically interesting. 🤖🏭

Not in a demo stage.
Not doing backflips.

Doing repetitive industrial work beside humans.

The real breakthrough for humanoid robotics will come when machines can reliably handle tools, materials and production tasks designed for human workers — without factories needing to rebuild everything around them.

That transition could redefine manufacturing.

Would you trust humanoid robots on a production line? 👇

#HumanoidRobots #Robotics #AI #Automation #5G #IoT #NLP #Tech #Technology #innovation #Bigdata

Cc: @jblefevre60 @Nicochan33 @pierrepinna @RosyCoaching @kalydeoo @Ym78200 @enilev @FrRonconii @CurieuxExplorer @Khulood_Almani @mvollmer1 @JoannMoretti @margaretsiegien @Shi4Tech @anand_narang @Eli_Krumova @HaroldSinnott @bimedotcom @EvanKirstel @Hana_ElSayyed @sallyeaves @SpirosMargaris @ipfconline1 @LaurentAlaus @Analytics_699 @Jeffkagan @AndresVilarino @efipm @FMFrancoise @MHcommunicate @Corix_JC @Archonsec @JolaBurnett @JeromeMONANGE @mary_gambara @drsharwood @smoothsale @sulefati7 @debashis_dutta @dinisguarda @devbhatt72 @terence_mills @theomitsa @BetaMoroney