RH20T
RH20T is a large-scale real-world robotic manipulation dataset of over 110,000 contact-rich manipulation sequences spanning 147 tasks (48 from RLBench, 29 from MetaWorld, 70 self-proposed), each paired with a human…
Source notes
RH20T is a large-scale real-world robotic manipulation dataset of over 110,000 contact-rich manipulation sequences spanning 147 tasks (48 from RLBench, 29 from MetaWorld, 70 self-proposed), each paired with a human demonstration video. Data was collected on 7 hardware configurations using four robot arms (Flexiv, UR5, Franka, Kuka) with 8-10 global RGBD cameras and 1-2 in-hand cameras per platform, capturing RGB, depth, binocular IR, joint angle/torque, gripper/EE pose, 6-DoF force-torque, audio, and fingertip tactile (Cfg7 only). The original data is ~40TB; the project provides a 640x360 resized version (~5TB RGB, ~10TB RGBD).
- Scale
- 110000 episodes
- Formats
- custom
- License
- custom
- Published
- 2023-07-02
Decision summary
human_demo
Metadata requires review against the official source before publication.
Read the dataset manifest and feature schema.
Selection readiness
6 evidence dimensions for deciding whether this dataset is ready to inspect, compare, or adopt. This is not a model benchmark.
Catalog evidence · not task performance
77/100
Provisional
4/6 dimensions scored · 48% confidence
Read the evidence behind all 6 dimensions
Access and governance
Access and license are declared by the source.
Schema and signal coverage
No machine-readable schema has been verified yet.
Policy training readiness
Observation and action/state signals are declared; alignment quality still depends on sample verification.
World-model readiness
Temporal observations plus geometry or semantic context are declared; sample alignment remains to be audited.
Failure and recovery readiness
No verified failure/recovery annotation evidence is available yet.
Download and processing readiness
Scale is declared; transfer and processing estimates are not measured.
Review unresolved evidence and next checks
Signal gaps
- Gaze / attention. Not enough evidence is available to classify this signal.
- Feedback / correction / failure. Not enough evidence is available to classify this signal.
Next checks
- Read the dataset manifest and feature schema.
- Run a bounded sample audit before assigning Strong readiness.
Dataset facts
- Source
- Shanghai Jiao Tong University (Cewu Lu Lab / MVIG)
- Evidence
- secondary claim
- Formats
- custom
- episodes
- 110000
- tasks
- 147
- bytes
- 40000000000000
Loop signals
5/7 present or partial
No decision-grade evidence captured yet.
unknownNo decision-grade evidence captured yet.
unknownView 5 more signal categories
depth · video · RH20T is a large-scale real-world robotic manipulation dataset of over 110,000 contact-rich manipulation sequences spanning 147 tasks (48 from RLBench, 29 from MetaWorld, 70 self-proposed), each paired with a human demonstration video. Data was collected on 7 hardware configurations using four robot arms (Flexiv, UR5, Franka, Kuka) with 8-10 global RGBD cameras and 1-2 in-hand cameras per platform, capturing RGB, depth, binocular IR, joint angle/torque, gripper/EE pose, 6-DoF force-torque, audio, and fingertip tactile (Cfg7 only). The original data is ~40TB; the project provides a 640x360 resized version (~5TB RGB, ~10TB RGBD).
presentee_pose · RH20T is a large-scale real-world robotic manipulation dataset of over 110,000 contact-rich manipulation sequences spanning 147 tasks (48 from RLBench, 29 from MetaWorld, 70 self-proposed), each paired with a human demonstration video. Data was collected on 7 hardware configurations using four robot arms (Flexiv, UR5, Franka, Kuka) with 8-10 global RGBD cameras and 1-2 in-hand cameras per platform, capturing RGB, depth, binocular IR, joint angle/torque, gripper/EE pose, 6-DoF force-torque, audio, and fingertip tactile (Cfg7 only). The original data is ~40TB; the project provides a 640x360 resized version (~5TB RGB, ~10TB RGBD).
partiallanguage · RH20T is a large-scale real-world robotic manipulation dataset of over 110,000 contact-rich manipulation sequences spanning 147 tasks (48 from RLBench, 29 from MetaWorld, 70 self-proposed), each paired with a human demonstration video. Data was collected on 7 hardware configurations using four robot arms (Flexiv, UR5, Franka, Kuka) with 8-10 global RGBD cameras and 1-2 in-hand cameras per platform, capturing RGB, depth, binocular IR, joint angle/torque, gripper/EE pose, 6-DoF force-torque, audio, and fingertip tactile (Cfg7 only). The original data is ~40TB; the project provides a 640x360 resized version (~5TB RGB, ~10TB RGBD).
presentdepth · RH20T is a large-scale real-world robotic manipulation dataset of over 110,000 contact-rich manipulation sequences spanning 147 tasks (48 from RLBench, 29 from MetaWorld, 70 self-proposed), each paired with a human demonstration video. Data was collected on 7 hardware configurations using four robot arms (Flexiv, UR5, Franka, Kuka) with 8-10 global RGBD cameras and 1-2 in-hand cameras per platform, capturing RGB, depth, binocular IR, joint angle/torque, gripper/EE pose, 6-DoF force-torque, audio, and fingertip tactile (Cfg7 only). The original data is ~40TB; the project provides a 640x360 resized version (~5TB RGB, ~10TB RGBD).
presentOpen · custom · custom
partialEvidence details and provenance
Official signal claims
Schema and annotations
No machine-readable schema facts are captured.
Sample verification
Pending. Metadata does not prove sample coverage, alignment, or file integrity.
curated official source
Integration notes
- Metadata requires review against the official source before publication.