Assets and datasets for RobotLearningLab manipulation tasks, covering scenarios from Sim2Lab, Real2Lab, and benchmarking. The dataset is authored by china-sae-robotics and was last updated on January 12, 2026. It is intended for use with RobotLearningLab for synthetic motion generation, policy training, and evaluation.
Use Cases
- Synthetic manipulation motion generation (SMMG) based on assets for manipulation tasks
- Imitation learning (IL) policy training based on manipulation task scenarios
- Vision-language-action (VLA) post-training based on multimodal assets
- Close-loop evaluation and deployment based on manipulation benchmarking scenarios
Strengths
- Assets cover multiple application scenarios including Sim2Lab and Real2Lab
- Supports a range of robotics development tasks from motion generation to policy deployment
Limitations
- Description metadata is limited; actual data quality requires manual inspection after download
- Column-level documentation is absent; field semantics must be inferred after download
- Row count is unknown, which may limit suitability assessment
Provenance
- Source
- china-sae-robotics
- Freshness
- Last updated 2026-01-12 02:25:15; freshness should be verified