Roborock: Robot Telemetry for Imitation Learning

Roborock is a dataset created using the LeRobot framework. It contains 3 episodes and 1,797 frames of robot telemetry data, likely for training imitation learning models. The dataset was uploaded by Yuan111111115 and last updated in June 2026.

Use Cases

• Train imitation learning models based on the robot telemetry data.
• Benchmark robot control algorithms based on the provided episodes and frames.
• Develop visual servoing policies based on the likely inclusion of video data.
• Analyze robot state transitions based on the sequential episode structure.

Strengths

• Contains 1,797 frames of robot data across 3 episodes.
• Data was created using the LeRobot framework, suggesting a structured collection process.
• Includes video data with a specified frame rate of 30 fps.

Limitations

• The dataset is relatively small, with only 3 episodes and 1 task.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Created using the LeRobot framework.

Collection Method

Likely contains robot telemetry and video data for imitation learning.

Freshness

Last updated 2026-06-10 12:09:49; freshness should be verified.