$π_0$: A Vision-Language-Action Flow Model for General Robot Control
This work addresses the problem of data, generalization, and robustness in robot learning for real-world systems, representing an incremental advancement in generalist robot policies.
The paper tackles the challenge of developing generalist robot policies for complex and dexterous tasks by proposing a flow matching architecture built on a pre-trained vision-language model, achieving zero-shot task performance and skill acquisition via fine-tuning across tasks like laundry folding and box assembly.
Robot learning holds tremendous promise to unlock the full potential of flexible, general, and dexterous robot systems, as well as to address some of the deepest questions in artificial intelligence. However, bringing robot learning to the level of generality required for effective real-world systems faces major obstacles in terms of data, generalization, and robustness. In this paper, we discuss how generalist robot policies (i.e., robot foundation models) can address these challenges, and how we can design effective generalist robot policies for complex and highly dexterous tasks. We propose a novel flow matching architecture built on top of a pre-trained vision-language model (VLM) to inherit Internet-scale semantic knowledge. We then discuss how this model can be trained on a large and diverse dataset from multiple dexterous robot platforms, including single-arm robots, dual-arm robots, and mobile manipulators. We evaluate our model in terms of its ability to perform tasks in zero shot after pre-training, follow language instructions from people and from a high-level VLM policy, and its ability to acquire new skills via fine-tuning. Our results cover a wide variety of tasks, such as laundry folding, table cleaning, and assembling boxes.