Ville🤖
2K posts
@VilleKuosmanen
gentleman scientist 🤖 @voyagerobotics
"RL Token" looks like a great and surprisingly simple post-training methodology for optimising robot models for dexterous tasks in the real world! Over the next few weeks, me and @pravsels will be attempting to reproduce the results (& open source the code) Stay tuned 👀
For serious winos, you can now buy a bucket of wine from Costco.
The robot is sorting and placing labels face-down at an impressive rate of ~23 packages/minute. It does make mistakes every now and then.
Unitree Unveils: GD01, A Manned Transformable Mecha, from $650,000 👏 The world's first production-ready manned mecha. It can transform. It's a civilian vehicle. It weighs ~500kg with you inside. Please everyone be sure to use the robot in a Friendly and Safe manner.
We have news! We created a new robotics model called Loop Model 1. On the zip-tie insertion task, it achieves 20x more throughput per unit of data than "Pi06 + RLT" from Physical Intelligence, a top model for such tasks. It’s the missing piece that makes MicroFactory work, because now deployment becomes so simple and fast that our users can do it themselves.
We just merged a small change to @LeRobotHF's default Diffusion Policy hyperparams, which should make it work better out of the box with real robots. If you've been using LeRobot to train mostly ACT policies, give a go to Diffusion Policy!
elder care is probably the biggest labor crisis in the world. where are the robots?"
🤖Low-data post-training can teach a VLA policy a new robot skill. But it also makes it too attached to the training demos. We call this lock-in🔒: the policy can execute the post-training task, yet fails to respond to seemingly obvious prompt changes. DeLock preserves steerability using only the policy’s own pretrained knowledge. No extra supervision needed!🚀🚀🚀 #Robotics #AI #EmbodiedAI #VLA
