Robotics

The π₀ release introduces a VLA generalist model for dexterous tasks like laundry folding and table bussing. π₀ uses a transformer with flow matching, combining VLM pre-training benefits and continuous action chunks at 50Hz, and is pre-trained on a broad dataset. With distinct pre-training and post-training stages, it supports zero-shot and fine-tuned task adaptation, demonstrating robustness to external interventions, as seen in an uncut video of π₀ folding laundry with a single model.

π₀ and its smaller, non-VLM version are evaluated against:

- Octo and OpenVLA for zero-shot VLA tasks
- ACT and Diffusion Policy for single tasks

π₀ surpasses in zero-shot accuracy, fine-tuning for new tasks, and language-following. Compute-parity ablations highlight trade-offs between VLA backbone gains and pre-training costs. Hierarchical methods like RT-H aid complex tasks needing low-level control and high-level planning, though Pi_0’s robust architecture largely drives its performance.

(link below)

creating and destroying; built flying cars & autonomous vehicles, now building roc.camera, writings: july.rocks

What a game changer, especially for people with physical disabilities,  paralysis, and the elderly. Hopefully at some point this technology will first become available to people who need it the most. I'm guessing Japan will be in early adopter. If it's ever mainstream in US I could definitely use a hand with laundry. 😅