Xarm Push Medium Replay Image: Robot Manipulation Demonstrations

LeRobot created this dataset for robot learning. It contains 800 episodes and 20,000 frames of data, recorded at 15 frames per second. The dataset was last updated on June 8, 2026.

Use Cases

• Training offline reinforcement learning agents based on recorded robot demonstrations.
• Developing imitation learning policies based on visual observation data.
• Benchmarking robot manipulation performance based on the Xarm robot platform.
• Analyzing robot behavior replay based on the medium replay strategy.

Strengths

• 800 episodes provide a substantial number of task demonstrations.
• 20,000 total frames offer a sequence of visual observations.
• Dataset structure includes defined splits for training.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data may reflect bias inherent to the specific robot platform and task setup.

Provenance

Source

LeRobot

Collection Method

Created using the LeRobot framework, likely involving recording robot demonstrations.

Freshness

Last updated 2026-06-08 18:24:09; freshness should be verified.