Datasets:

RoboCOIN
/

AIRBOT_MMK2_stacking_blocks

task_categories:
  - robotics
language:
  - en
  - zh
extra_gated_prompt: >-
  By accessing this dataset, you agree to cite the associated paper in your
  research/publications—see the "Citation" section for details. You agree to not
  use the dataset to conduct experiments that cause harm to human subjects.
extra_gated_fields:
  Company/Organization:
    type: text
    description: e.g., "ETH Zurich", "Boston Dynamics", "Independent Researcher"
  Country:
    type: country
    description: e.g., "Germany", "China", "United States"
tags:
  - RoboCOIN
  - LeRobot
frame_range: 10K-100K
license: apache-2.0
configs:
  - config_name: default
    data_files: data/*/*.parquet

AIRBOT_MMK2_stacking_blocks

📋 Overview

This dataset uses an extended format based on LeRobot and is fully compatible with LeRobot.

Robot Type: discover_robotics_aitbot_mmk2 | Codebase Version: v2.1 End-Effector Type: five_finger_hand

🏠 Scene Types

This dataset covers the following scene types:

home

🤖 Atomic Actions

This dataset includes the following atomic actions:

grasp
pick
place

📊 Dataset Statistics

Metric	Value
Total Episodes	213
Total Frames	52766
Total Tasks	5
Total Videos	852
Total Chunks	1
Chunk Size	1000
FPS	30
Dataset Size	2.2GB

👥 Authors

Contributors

This dataset is contributed by:

RoboCOIN - RoboCOIN Team

🔗 Links

🏠 Homepage: https://flagopen.github.io/RoboCOIN/
📄 Paper: https://arxiv.org/abs/2511.17441
💻 Repository: https://github.com/FlagOpen/RoboCOIN
🌐 Project Page: https://flagopen.github.io/RoboCOIN/
🐛 Issues: https://github.com/FlagOpen/RoboCOIN/issues
📜 License: apache-2.0

🏷️ Dataset Tags

RoboCOIN
LeRobot

🎯 Task Descriptions

Primary Tasks

Pick up the red rectangular block from one side of the table with one hand and place it in the center, then pick up the blue hexagonal wooden block from the other side with your other hand and place it on top of the red block. Move the large block from one side to the center with one hand, and stack the small blocks on top of it with your other hand. Pick up a block from the table with one hand and place it on the blocks, then pick up another block from the table with your other hand and place it on the blocks. Pick up the blue cube from one side of the table with one hand and place it on top of the yellow cube in the middle, then pick up the green cube from the other side with your other hand and place it on top of the blue cube. Pick up the blue cube from the table with one hand and place it on the small table, then pick up the red cube from the table with your other hand and place it on top of the blue cube.

Sub-Tasks

This dataset includes 25 distinct subtasks:

Place the yellow cube on the orange cube with the right gripper
Static
Grasp the blue rhombic cuboid block with the right gripper
Grasp the red cuboid block with the left gripper
Grasp the blue small cube with the left gripper
Place the red small cube on the blue small cube with the right gripper
Place the red cuboid block in the center of view with the left gripper
Grasp the big red cube with the left gripper
Abnormal
Place the orange cube on the yellow cube with the right gripper
Place the green small cube on the blue small cube with the right gripper
Grasp the green small cube with the right gripper
Grasp the red small cube with the right gripper
Grasp the blue cube with the right gripper
Place the blue cube on the red cube with the left gripper
Grasp the orange cube with the right gripper
Grasp the blue cube with the left gripper
Place the blue small cube on the small table with the left gripper
End
Place the blue rhombic cuboid block on the red cuboid block with the right gripper
Place the blue small cube on the yellow small cube with the left gripper
Place the big red cube in the center of view with the left gripper
Grasp the yellow cube with the right gripper
Place theblue cube on the big red cube with the right gripper
null

🎥 Camera Views

This dataset includes 4 camera views.

🏷️ Available Annotations

This dataset includes rich annotations to support diverse learning approaches:

Subtask Annotations

Subtask Segmentation: Fine-grained subtask segmentation and labeling

Scene Annotations

Scene-level Descriptions: Semantic scene classifications and descriptions

End-Effector Annotations

Direction: Movement direction classifications for robot end-effectors
Velocity: Velocity magnitude categorizations during manipulation
Acceleration: Acceleration magnitude classifications for motion analysis

Gripper Annotations

Gripper Mode: Open/close state annotations for gripper control
Gripper Activity: Activity state classifications (active/inactive)

Additional Features

End-Effector Simulation Pose: 6D pose information for end-effectors in simulation space
- Available for both state and action
Gripper Opening Scale: Continuous gripper opening measurements
- Available for both state and action

📂 Data Splits

The dataset is organized into the following splits:

Training: Episodes 0:212

📁 Dataset Structure

This dataset follows the LeRobot format and contains the following components:

Data Files

Videos: Compressed video files containing RGB camera observations
State Data: Robot joint positions, velocities, and other state information
Action Data: Robot action commands and trajectories
Metadata: Episode metadata, timestamps, and annotations

File Organization

Data Path Pattern: data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet
Video Path Pattern: videos/chunk-{episode_chunk:03d}/{video_key}/episode_{episode_index:06d}.mp4
Chunking: Data is organized into 1 chunk(s) of size 1000

Features Schema

The dataset includes the following features:

Visual Observations

observation.images.cam_high_rgb: video
- FPS: 30
- Codec: av1- observation.images.cam_left_wrist_rgb: video
- FPS: 30
- Codec: av1- observation.images.cam_right_wrist_rgb: video
- FPS: 30
- Codec: av1- observation.images.cam_third_view: video
- FPS: 30
- Codec: av1

State and Action- observation.state: float32- action: float32

Temporal Information

timestamp: float32
frame_index: int64
episode_index: int64
index: int64
task_index: int64

Annotations

subtask_annotation: int32
scene_annotation: int32

Motion Features

eef_sim_pose_state: float32
- Dimensions: left_eef_pos_x, left_eef_pos_y, left_eef_pos_z, left_eef_ori_x, left_eef_ori_y, left_eef_ori_z, right_eef_pos_x, right_eef_pos_y, right_eef_pos_z, right_eef_ori_x, right_eef_ori_y, right_eef_ori_z
eef_sim_pose_action: float32
- Dimensions: left_eef_pos_x, left_eef_pos_y, left_eef_pos_z, left_eef_ori_x, left_eef_ori_y, left_eef_ori_z, right_eef_pos_x, right_eef_pos_y, right_eef_pos_z, right_eef_ori_x, right_eef_ori_y, right_eef_ori_z
eef_direction_state: int32
- Dimensions: left_eef_direction, right_eef_direction
eef_direction_action: int32
- Dimensions: left_eef_direction, right_eef_direction
eef_velocity_state: int32
- Dimensions: left_eef_velocity, right_eef_velocity
eef_velocity_action: int32
- Dimensions: left_eef_velocity, right_eef_velocity
eef_acc_mag_state: int32
- Dimensions: left_eef_acc_mag, right_eef_acc_mag
eef_acc_mag_action: int32
- Dimensions: left_eef_acc_mag, right_eef_acc_mag

Gripper Features

Meta Information

The complete dataset metadata is available in meta/info.json:

{"codebase_version": "v2.1", "robot_type": "discover_robotics_aitbot_mmk2", "total_episodes": 213, "total_frames": 52766, "total_tasks": 5, "total_videos": 852, "total_chunks": 1, "chunks_size": 1000, "fps": 30, "splits": {"train": "0:212"}, "data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet", "video_path": "videos/chunk-{episode_chunk:03d}/{video_key}/episode_{episode_index:06d}.mp4", "features": {"observation.images.cam_high_rgb": {"dtype": "video", "shape": [480, 640, 3], "names": ["height", "width", "channels"], "info": {"video.height": 480, "video.width": 640, "video.codec": "av1", "video.pix_fmt": "yuv420p", "video.is_depth_map": false, "video.fps": 30, "video.channels": 3, "has_audio": false}}, "observation.images.cam_left_wrist_rgb": {"dtype": "video", "shape": [480, 640, 3], "names": ["height", "width", "channels"], "info": {"video.height": 480, "video.width": 640, "video.codec": "av1", "video.pix_fmt": "yuv420p", "video.is_depth_map": false, "video.fps": 30, "video.channels": 3, "has_audio": false}}, "observation.images.cam_right_wrist_rgb": {"dtype": "video", "shape": [480, 640, 3], "names": ["height", "width", "channels"], "info": {"video.height": 480, "video.width": 640, "video.codec": "av1", "video.pix_fmt": "yuv420p", "video.is_depth_map": false, "video.fps": 30, "video.channels": 3, "has_audio": false}}, "observation.images.cam_third_view": {"dtype": "video", "shape": [480, 640, 3], "names": ["height", "width", "channels"], "info": {"video.height": 480, "video.width": 640, "video.codec": "av1", "video.pix_fmt": "yuv420p", "video.is_depth_map": false, "video.fps": 30, "video.channels": 3, "has_audio": false}}, "observation.state": {"dtype": "float32", "shape": [36], "names": ["left_arm_joint_1_rad", "left_arm_joint_2_rad", "left_arm_joint_3_rad", "left_arm_joint_4_rad", "left_arm_joint_5_rad", "left_arm_joint_6_rad", "right_arm_joint_1_rad", "right_arm_joint_2_rad", "right_arm_joint_3_rad", "right_arm_joint_4_rad", "right_arm_joint_5_rad", "right_arm_joint_6_rad", "left_hand_joint_1_rad", "left_hand_joint_2_rad", "left_hand_joint_3_rad", "left_hand_joint_4_rad", "left_hand_joint_5_rad", "left_hand_joint_6_rad", "left_hand_joint_7_rad", "left_hand_joint_8_rad", "left_hand_joint_9_rad", "left_hand_joint_10_rad", "left_hand_joint_11_rad", "left_hand_joint_12_rad", "right_hand_joint_1_rad", "right_hand_joint_2_rad", "right_hand_joint_3_rad", "right_hand_joint_4_rad", "right_hand_joint_5_rad", "right_hand_joint_6_rad", "right_hand_joint_7_rad", "right_hand_joint_8_rad", "right_hand_joint_9_rad", "right_hand_joint_10_rad", "right_hand_joint_11_rad", "right_hand_joint_12_rad"]}, "action": {"dtype": "float32", "shape": [36], "names": ["left_arm_joint_1_rad", "left_arm_joint_2_rad", "left_arm_joint_3_rad", "left_arm_joint_4_rad", "left_arm_joint_5_rad", "left_arm_joint_6_rad", "right_arm_joint_1_rad", "right_arm_joint_2_rad", "right_arm_joint_3_rad", "right_arm_joint_4_rad", "right_arm_joint_5_rad", "right_arm_joint_6_rad", "left_hand_joint_1_rad", "left_hand_joint_2_rad", "left_hand_joint_3_rad", "left_hand_joint_4_rad", "left_hand_joint_5_rad", "left_hand_joint_6_rad", "left_hand_joint_7_rad", "left_hand_joint_8_rad", "left_hand_joint_9_rad", "left_hand_joint_10_rad", "left_hand_joint_11_rad", "left_hand_joint_12_rad", "right_hand_joint_1_rad", "right_hand_joint_2_rad", "right_hand_joint_3_rad", "right_hand_joint_4_rad", "right_hand_joint_5_rad", "right_hand_joint_6_rad", "right_hand_joint_7_rad", "right_hand_joint_8_rad", "right_hand_joint_9_rad", "right_hand_joint_10_rad", "right_hand_joint_11_rad", "right_hand_joint_12_rad"]}, "timestamp": {"dtype": "float32", "shape": [1], "names": null}, "frame_index": {"dtype": "int64", "shape": [1], "names": null}, "episode_index": {"dtype": "int64", "shape": [1], "names": null}, "index": {"dtype": "int64", "shape": [1], "names": null}, "task_index": {"dtype": "int64", "shape": [1], "names": null}, "subtask_annotation": {"names": null, "dtype": "int32", "shape": [5]}, "scene_annotation": {"names": null, "dtype": "int32", "shape": [1]}, "eef_sim_pose_state": {"names": ["left_eef_pos_x", "left_eef_pos_y", "left_eef_pos_z", "left_eef_ori_x", "left_eef_ori_y", "left_eef_ori_z", "right_eef_pos_x", "right_eef_pos_y", "right_eef_pos_z", "right_eef_ori_x", "right_eef_ori_y", "right_eef_ori_z"], "dtype": "float32", "shape": [12]}, "eef_sim_pose_action": {"names": ["left_eef_pos_x", "left_eef_pos_y", "left_eef_pos_z", "left_eef_ori_x", "left_eef_ori_y", "left_eef_ori_z", "right_eef_pos_x", "right_eef_pos_y", "right_eef_pos_z", "right_eef_ori_x", "right_eef_ori_y", "right_eef_ori_z"], "dtype": "float32", "shape": [12]}, "eef_direction_state": {"names": ["left_eef_direction", "right_eef_direction"], "dtype": "int32", "shape": [2]}, "eef_direction_action": {"names": ["left_eef_direction", "right_eef_direction"], "dtype": "int32", "shape": [2]}, "eef_velocity_state": {"names": ["left_eef_velocity", "right_eef_velocity"], "dtype": "int32", "shape": [2]}, "eef_velocity_action": {"names": ["left_eef_velocity", "right_eef_velocity"], "dtype": "int32", "shape": [2]}, "eef_acc_mag_state": {"names": ["left_eef_acc_mag", "right_eef_acc_mag"], "dtype": "int32", "shape": [2]}, "eef_acc_mag_action": {"names": ["left_eef_acc_mag", "right_eef_acc_mag"], "dtype": "int32", "shape": [2]}}}

Directory Structure

The dataset is organized as follows (showing leaf directories with first 5 files only):

AIRBOT_MMK2_stacking_blocks_qced_hardlink/
├── annotations/
│   ├── eef_acc_mag_annotation.jsonl
│   ├── eef_direction_annotation.jsonl
│   ├── eef_velocity_annotation.jsonl
│   ├── gripper_activity_annotation.jsonl
│   ├── gripper_mode_annotation.jsonl
│   └── (...)
├── data/
│   └── chunk-000/
│       ├── episode_000000.parquet
│       ├── episode_000001.parquet
│       ├── episode_000002.parquet
│       ├── episode_000003.parquet
│       ├── episode_000004.parquet
│       └── (...)
├── meta/
│   ├── episodes.jsonl
│   ├── episodes_stats.jsonl
│   ├── info.json
│   └── tasks.jsonl
└── videos/
    └── chunk-000/
        ├── observation.images.cam_high_rgb/
        │   ├── episode_000000.mp4
        │   ├── episode_000001.mp4
        │   ├── episode_000002.mp4
        │   ├── episode_000003.mp4
        │   ├── episode_000004.mp4
        │   └── (...)
        ├── observation.images.cam_left_wrist_rgb/
        │   ├── episode_000000.mp4
        │   ├── episode_000001.mp4
        │   ├── episode_000002.mp4
        │   ├── episode_000003.mp4
        │   ├── episode_000004.mp4
        │   └── (...)
        ├── observation.images.cam_right_wrist_rgb/
        │   ├── episode_000000.mp4
        │   ├── episode_000001.mp4
        │   ├── episode_000002.mp4
        │   ├── episode_000003.mp4
        │   ├── episode_000004.mp4
        │   └── (...)
        └── observation.images.cam_third_view/
            ├── episode_000000.mp4
            ├── episode_000001.mp4
            ├── episode_000002.mp4
            ├── episode_000003.mp4
            ├── episode_000004.mp4
            └── (...)

📞 Contact and Support

For questions, issues, or feedback regarding this dataset, please contact:

Email: None For questions, issues, or feedback regarding this dataset, please contact us.

Support

For technical support, please open an issue on our GitHub repository.

📄 License

This dataset is released under the apache-2.0 license.

Please refer to the LICENSE file for full license terms and conditions.

📚 Citation

If you use this dataset in your research, please cite:

@article{robocoin,
    title={RoboCOIN: An Open-Sourced Bimanual Robotic Data Collection for Integrated Manipulation},
    author={Shihan Wu, Xuecheng Liu, Shaoxuan Xie, Pengwei Wang, Xinghang Li, Bowen Yang, Zhe Li, Kai Zhu, Hongyu Wu, Yiheng Liu, Zhaoye Long, Yue Wang, Chong Liu, Dihan Wang, Ziqiang Ni, Xiang Yang, You Liu, Ruoxuan Feng, Runtian Xu, Lei Zhang, Denghang Huang, Chenghao Jin, Anlan Yin, Xinlong Wang, Zhenguo Sun, Junkai Zhao, Mengfei Du, Mingyu Cao, Xiansheng Chen, Hongyang Cheng, Xiaojie Zhang, Yankai Fu, Ning Chen, Cheng Chi, Sixiang Chen, Huaihai Lyu, Xiaoshuai Hao, Yequan Wang, Bo Lei, Dong Liu, Xi Yang, Yance Jiao, Tengfei Pan, Yunyan Zhang, Songjing Wang, Ziqian Zhang, Xu Liu, Ji Zhang, Caowei Meng, Zhizheng Zhang, Jiyang Gao, Song Wang, Xiaokun Leng, Zhiqiang Xie, Zhenzhen Zhou, Peng Huang, Wu Yang, Yandong Guo, Yichao Zhu, Suibing Zheng, Hao Cheng, Xinmin Ding, Yang Yue, Huanqian Wang, Chi Chen, Jingrui Pang, YuXi Qian, Haoran Geng, Lianli Gao, Haiyuan Li, Bin Fang, Gao Huang, Yaodong Yang, Hao Dong, He Wang, Hang Zhao, Yadong Mu, Di Hu, Hao Zhao, Tiejun Huang, Shanghang Zhang, Yonghua Lin, Zhongyuan Wang and Guocai Yao},
    journal={arXiv preprint arXiv:2511.17441},
    url = {https://arxiv.org/abs/2511.17441},
    year={2025}
    }

Additional References

If you use this dataset, please also consider citing:

LeRobot Framework: https://github.com/huggingface/lerobot

📌 Version Information

Version History

v1.0.0 (2025-11): Initial release