RoboNet
by Berkeley AI Research (BAIR) · free · Last verified 2026-03-17
RoboNet is a large-scale dataset for robot learning, featuring 15 million video frames from diverse robot arms across multiple labs. It is designed to train and benchmark self-supervised visual models, aiming to achieve generalization across different robot morphologies and workspaces without task-specific labels.
https://www.robonet.wiki ↗B
B—Above Average
Adoption: BQuality: AFreshness: BCitations: B+Engagement: F
Specifications
- License
- CC BY 4.0
- Pricing
- free
- Capabilities
- visual-imitation-learning, self-supervised-representation-learning, video-prediction-models, cross-robot-policy-transfer, reinforcement-learning-from-demonstrations, robot-manipulation-benchmarking, visual-foresight-training
- Integrations
- Use Cases
- [object Object], [object Object], [object Object], [object Object]
- API Available
- No
- Tags
- robotics, video, manipulation, multi-robot, self-supervised, dataset, imitation-learning, reinforcement-learning, computer-vision, robot-learning, deep-learning
- Added
- 2026-03-17
- Completeness
- 0.6%
Index Score
60.3Adoption
62
Quality
80
Freshness
65
Citations
78
Engagement
0