BridgeData V2: A Dataset for Robot Learning at Scale
This provides a valuable resource for researchers in robotics and AI to accelerate work on scalable robot learning, though it is incremental as it builds on existing dataset efforts.
The authors tackled the challenge of scalable robot learning by introducing BridgeData V2, a large and diverse dataset of 60,096 robotic manipulation trajectories collected across 24 environments, which enabled training of 6 state-of-the-art methods that showed improved performance and generalization with more data and variety.
We introduce BridgeData V2, a large and diverse dataset of robotic manipulation behaviors designed to facilitate research on scalable robot learning. BridgeData V2 contains 60,096 trajectories collected across 24 environments on a publicly available low-cost robot. BridgeData V2 provides extensive task and environment variability, leading to skills that can generalize across environments, domains, and institutions, making the dataset a useful resource for a broad range of researchers. Additionally, the dataset is compatible with a wide variety of open-vocabulary, multi-task learning methods conditioned on goal images or natural language instructions. In our experiments, we train 6 state-of-the-art imitation learning and offline reinforcement learning methods on our dataset, and find that they succeed on a suite of tasks requiring varying amounts of generalization. We also demonstrate that the performance of these methods improves with more data and higher capacity models, and that training on a greater variety of skills leads to improved generalization. By publicly sharing BridgeData V2 and our pre-trained models, we aim to accelerate research in scalable robot learning methods. Project page at https://rail-berkeley.github.io/bridgedata