Cool! By isolating the actual mechanisms behind generative control performance, this paper encourages a much-needed rethinking of modern robot policy design.
Generative models (diffusion/flow) are taking over robotics 🤖. But do we really need to model the full action distribution to control a robot?
We suspected the success of Generative Control Policies (GCPs) might be "Much Ado About Noising."
We rigorously tested the myths. 🧵👇
Pixels in, contacts out...
Perception, interaction, autonomy - next agenda for humanoids.
We learn a multi-task humanoid world model from offline datasets and use MPC to plan contact-aware behaviors from ego-vision in the real-world.
Project and Code: ego-vcp.github.io
