About the Project This was a global hackathon that I and some of my friends from the lab joined. It was about the use of LeRobot's SO-101 robots and Huggingface's new foundation model, a Vision-Language-Action (VLA) called SmolVLA . We used it for a multi-robot collaboration task where two of these robots come together to pick up a specific object and transfer between them. Two SmolVLA models run together and we fine-tuned them using imitation learning for our this task. We