Iván López Broceño - MSc Robotics Engineer @TUDelft - Software Robotics Engineering Bachelor @URJC
What I cannot create, I do not understand - R. Feynman
Introducing ⚡️Flash ALOHA: Learning Fast and Precise Manipulation!
With just 50 demonstrations, Flash ALOHA learns to autonomously complete dynamic and precise tasks at near human speed.
Everything open sourced!
(1/11)
Team:
@Qi_Wu577@YihuaiGao@chelseabfinn@zipengfu
Because I'm hiring a bunch of new PhD students, I took some time to think about what it means to start a research career in the age of competent AI agents. And then I wrote another long blog post. I'm not pompous enough to call it an essay, but you may want to read it anyway.
Can a robot manipulate a rope, granular objects or cloth to reach precise 3D shapes?
🚨Introducing DefFlow: a goal-conditioned 3D World Action Modelling approach that learns to manipulate deformable objects into diverse goal shapes using 3D point flows.
A thread🧵(1/n)
🚨 BREAKING:
Meet AIR, a dual-arm humanoid built by @EmbodiedAI_ch for factory floors.
The company is an EPFL spin-off, headquartered in Lausanne, opening an office in Rome.
Founded by Francesco Stella, Kai Junge and Max Polzin, PhDs from EPFL, with backgrounds from MIT, ETH and Cambridge.
The company is part of Google DeepMind accelerator and NVIDIA Inception member.
The design choices tell you everything about their philosophy. Just have a look!
No legs. An omnidirectional wheeled base instead. A 100kg biped losing balance next to an operator is a safety incident. Factory floors are flat, legs don't buy you anything there.
Mass in the torso, not the arms. Light, tendon-driven arms mean no cage, and no fence required. The robot can stand next to a person on a production line.
AIR drops into a workstation that already exists, so there's no need for line rebuild (less investment into infra).
The deployment model is equally smart:
→ Productive from the first shift under remote operation
→ Learns the task from the work it performs
→ Becomes autonomous within months, teleoperation as fallback only
Built for high-mix industrial work, kitting, assembly, cabling, machine tending. The tasks that defeat rigid single-purpose automation.
Embodied AI is optimising for the constraint that actually decides whether a robot gets deployed: can it safely stand next to a human on a real factory floor?
Europe building its own industrial robot workforce! 🇪🇺
VERY INTERESTING.
Great to be onboard as one of the early investors! 🦾
~~
♻️ Join the weekly robotics newsletter, and never miss any news → https://t.co/GoA3ZuwoPB
Predicting what a robot will feel as it moves.
Our 3D World Simulator learns to predict touch as 3D particles alongside scene geometry, keeping tactile signals aligned with contact surfaces as interactions unfold. Learning from these predicted visual and tactile observations helps robots perform contact-rich manipulation in the real world.
Glad to see a world model learning from FlexiTac data. The tactile predictions look really impressive!
Just released the first interactive session (out of 4) of my Reinforcement Learning class, modestly titled "Modeling the world" 🎯
👉 Try it here:
https://t.co/5DeG5GuUvS
A robot duck walking through simulated snow! 🦆
MicroDuck from @huggingface walking through a snowy scenery simulated in NVIDIA Newton with a custom snow solver.
We all know that this robot duck is quite comfortable on the stage, but right now we can see it's quite capable in the snow too! ❄
The snow is done properly, and it's very important to go more in details here.
Flake sizes, masses and fall speeds come from measured snowfall literature going back to the 1970s. Flakes tumble like thin plates. The rate is a real 8 mm/h liquid equivalent, around 290,000 flakes airborne at once.
Snow sticks according to the angle of repose at air temperature, rides along in each link’s own frame as the robot moves, and the feet press footprints into a 2 mm-resolution snowpack.
The walking policy was trained in Isaac Lab and reimplemented in pure Newton, without the Isaac Lab runtime.
Then checked against Isaac Lab running headless: observations matching to 1e-6, actions to 2e-6, an identical servo model across 4,088 random inputs, walking speed within 0.003 m/s.
Twenty-two of twenty-two checks! ✅
Rendering snow is achievable!!!
@TheRobotSimGuy 🐐, @NVIDIARobotics
~~
♻️ Join the weekly robotics newsletter, and never miss any news → https://t.co/GoA3ZuwoPB
Ever wondered what it’s like to learn robot autonomy @Stanford? Now you can.
Today, we’re opening up the robot autonomy teaching materials that my collaborators and I have been developing over the past several years—for anyone, anywhere, completely free.
At the heart of this effort is Principles of Robot Autonomy (PoRA), a textbook inspired by the Stanford robot autonomy course series I’ve been teaching since 2017.
Robot autonomy has evolved from a research discipline into a technology deployed in the real world. Building capable autonomous systems requires more than understanding individual algorithms: it requires understanding how perception, state estimation, planning, control, and learning work together in a complete autonomy stack.
That’s what we set out to teach with PoRA: bringing classical robotics and modern AI together in one coherent framework, grounded in both theory and practice.
Coming soon in print from Cambridge University Press, Principles of Robot Autonomy is also freely available online and comes with a comprehensive collection of open educational resources:
📖 Textbook, https://t.co/wRvXr4VCIP — Foundations of robot autonomy, from classical methods to modern AI-based approaches.
🎥 Video lectures, https://t.co/Yw3P1aGzBG Lectures from my Stanford courses on robot autonomy and on optimal and learning-based control.
💻 Interactive exercises, https://t.co/kyPV8AnyB3 — Hands-on Python implementations and Jupyter notebooks.
📝 Teaching materials, https://t.co/a6MqqYAcEN — Lecture slides and resources for instructors.
🌐 Explore all PoRA resources, https://t.co/oAsjrMJxZT — Including Ask the Book, an AI study assistant integrated throughout the textbook to answer questions, explain derivations, and connect ideas across chapters.
📣 Read the Stanford announcement: https://t.co/f6BBBZZQmg
Our goal is simple: to make high-quality education in robot autonomy accessible to students, researchers, engineers, educators, and anyone curious about how autonomous robotic systems work.
I believe that sharing knowledge is fundamental to the mission of a university. Autonomous robotics—and Physical AI more broadly—are among the defining technological frontiers of our time. We want to help more people understand these technologies, build with them, and advance the field.
A huge thank you to my co-authors, @DanieleGammelli, Joseph Lorenzetti, @katielulula, and @GioeleZardini, and to everyone who helped bring this project to life.
Explore the resources, use them in your own learning or teaching, and please share them with anyone who might find them useful!
#Robotics #PhysicalAI #ArtificialIntelligence #AutonomousSystems #Education
@Stanford@StanfordEng@StanfordAILab@StanfordHAI
The @JensenHuang keynote effect on https://t.co/RR7kOJhWIW pre-orders.
Going to cross 20,000+ total pre-orders probably tomorrow making Microduck probably the most sold robot of 2026?