jnogga/droid_failure

---
license: cc-by-4.0
task_categories:
- robotics
tags:
- LeRobot
- Robotic manipulation
pretty_name: DROID Failure Episodes
size_categories:
- 1M<n<10M
---

## DROID Failure Episodes

Failure episodes in [DROID-COMMUNITY](https://droid-dataset.github.io/). Episodes with inconsistent data were discarded, leaving ~91% of all failure episodes.

Ported from raw 1.0.1 data at full resolution to LeRobotDataset v3.0 format (0.25 TiB | 1.5k inodes) with extra annotation from [KarlP/droid](https://huggingface.co/KarlP/droid/tree/main).

<div align="center" style="margin: 16px 0;">
<video controls autoplay loop muted playsinline style="max-width: 100%; border-radius: 10px;">
<source src="https://huggingface.co/datasets/jnogga/droid_failure/resolve/main/droid_example_episode.mp4" type="video/mp4">
Your browser does not support the video tag.
</video>
</div>

For the successful episodes, see [jnogga/droid_success](https://huggingface.co/datasets/jnogga/droid_success).

## Updates

* 22.02.2026: Removed episodes with unexpected camera resolutions.

## Dataset Structure

The external cameras are assigned to left and right views depending on the episode. For their extrinsics, the source is given along with the corresponding metric and score.

[meta/info.json](meta/info.json):
```json
{
"codebase_version": "v3.0",
"fps": 15,
"data_files_size_in_mb": 100.0,
"video_files_size_in_mb": 200.0,
"robot_type": "franka_emika_panda_robotiq_2F-85",
"total_episodes": 13747,
"total_frames": 3476037,
"total_tasks": 13747,
"chunks_size": 1000,
"splits": {
"train": "0:13747"
},
"data_path": "data/chunk-{chunk_index:03d}/file_{file_index:03d}.parquet",
"video_path": "videos/{video_key}/chunk-{chunk_index:03d}/file_{file_index:03d}.mp4",
"features": {
"action.cartesian_position": {
"dtype": "float32",
"shape": [
7
],
"names": [
"position.x",
"position.y",
"position.z",
"quaternion.w",
"quaternion.x",
"quaternion.y",
"quaternion.z"
],
"fps": 15
},
"action.cartesian_velocity": {
"dtype": "float32",
"shape": [
7
],
"names": [
"position.x",
"position.y",
"position.z",
"quaternion.w",
"quaternion.x",
"quaternion.y",
"quaternion.z"
],
"fps": 15
},
"action.gripper_position": {
"dtype": "float32",
"shape": [
1
],
"names": null,
"fps": 15
},
"action.gripper_velocity": {
"dtype": "float32",
"shape": [
1
],
"names": null,
"fps": 15
},
"action.joint_position": {
"dtype": "float32",
"shape": [
7
],
"names": [
"joint_0",
"joint_1",
"joint_2",
"joint_3",
"joint_4",
"joint_5",
"joint_6"
],
"fps": 15
},
"action.joint_velocity": {
"dtype": "float32",
"shape": [
7
],
"names": [
"joint_0",
"joint_1",
"joint_2",
"joint_3",
"joint_4",
"joint_5",
"joint_6"
],
"fps": 15
},
"observation.cartesian_position": {
"dtype": "float32",
"shape": [
7
],
"names": [
"position.x",
"position.y",
"position.z",
"quaternion.w",
"quaternion.x",
"quaternion.y",
"quaternion.z"
],
"fps": 15
},
"observation.gripper_position": {
"dtype": "float32",
"shape": [
1
],
"names": null,
"fps": 15
},
"observation.joint_position": {
"dtype": "float32",
"shape": [
7
],
"names": [
"joint_0",
"joint_1",
"joint_2",
"joint_3",
"joint_4",
"joint_5",
"joint_6"
],
"fps": 15
},
"observation.joint_velocity": {
"dtype": "float32",
"shape": [
7
],
"names": [
"joint_0",
"joint_1",
"joint_2",
"joint_3",
"joint_4",
"joint_5",
"joint_6"
],
"fps": 15
},
"wrist_camera_intrinsics": {
"dtype": "float32",
"shape": [
4
],
"names": [
"fx",
"cx",
"fy",
"cy"
],
"fps": 15
},
"wrist_camera_to_robot_extrinsics": {
"dtype": "float32",
"shape": [
7
],
"names": [
"position.x",
"position.y",
"position.z",
"quaternion.w",
"quaternion.x",
"quaternion.y",
"quaternion.z"
],
"fps": 15
},
"left_external_camera_intrinsics": {
"dtype": "float32",
"shape": [
4
],
"names": [
"fx",
"cx",
"fy",
"cy"
],
"fps": 15
},
"left_external_camera_to_robot_extrinsics": {
"dtype": "float32",
"shape": [
7
],
"names": [
"position.x",
"position.y",
"position.z",
"quaternion.w",
"quaternion.x",
"quaternion.y",
"quaternion.z"
],
"fps": 15
},
"left_external_camera_to_robot_source": {
"dtype": "string",
"shape": [
1
],
"names": null,
"fps": 15
},
"left_external_camera_to_robot_metric": {
"dtype": "string",
"shape": [
1
],
"names": null,
"fps": 15
},
"left_external_camera_to_robot_score": {
"dtype": "float32",
"shape": [
1
],
"names": null,
"fps": 15
},
"left_external_camera_to_robot_extrinsics_raw": {
"dtype": "float32",
"shape": [
7
],
"names": [
"position.x",
"position.y",
"position.z",
"quaternion.w",
"quaternion.x",
"quaternion.y",
"quaternion.z"
],
"fps": 15
},
"right_external_camera_intrinsics": {
"dtype": "float32",
"shape": [
4
],
"names": [
"fx",
"cx",
"fy",
"cy"
],
"fps": 15
},
"right_external_camera_to_robot_extrinsics": {
"dtype": "float32",
"shape": [
7
],
"names": [
"position.x",
"position.y",
"position.z",
"quaternion.w",
"quaternion.x",
"quaternion.y",
"quaternion.z"
],
"fps": 15
},
"right_external_camera_to_robot_source": {
"dtype": "string",
"shape": [
1
],
"names": null,
"fps": 15
},
"right_external_camera_to_robot_metric": {
"dtype": "string",
"shape": [
1
],
"names": null,
"fps": 15
},
"right_external_camera_to_robot_score": {
"dtype": "float32",
"shape": [
1
],
"names": null,
"fps": 15
},
"right_external_camera_to_robot_extrinsics_raw": {
"dtype": "float32",
"shape": [
7
],
"names": [
"position.x",
"position.y",
"position.z",
"quaternion.w",
"quaternion.x",
"quaternion.y",
"quaternion.z"
],
"fps": 15
},
"keep_interval_index": {
"dtype": "int64",
"shape": [
1
],
"names": null,
"fps": 15
},
"frame_index": {
"dtype": "int64",
"shape": [
1
],
"names": null,
"fps": 15
},
"timestamp": {
"dtype": "float32",
"shape": [
1
],
"names": null,
"fps": 15
},
"index": {
"dtype": "int64",
"shape": [
1
],
"names": null,
"fps": 15
},
"task_index": {
"dtype": "int64",
"shape": [
1
],
"names": null,
"fps": 15
},
"episode_index": {
"dtype": "int64",
"shape": [
1
],
"names": null,
"fps": 15
},
"observation.images.left_external": {
"dtype": "video",
"shape": [
720,
1280,
3
],
"names": [
"height",
"width",
"channel"
],
"info": {
"video.height": 720,
"video.width": 1280,
"video.codec": "h264",
"video.pix_fmt": "yuv420p",
"video.is_depth_map": false,
"video.fps": 15,
"video.channels": 3,
"has_audio": false
}
},
"observation.images.wrist": {
"dtype": "video",
"shape": [
720,
1280,
3
],
"names": [
"height",
"width",
"channel"
],
"info": {
"video.height": 720,
"video.width": 1280,
"video.codec": "h264",
"video.pix_fmt": "yuv420p",
"video.is_depth_map": false,
"video.fps": 15,
"video.channels": 3,
"has_audio": false
}
},
"observation.images.right_external": {
"dtype": "video",
"shape": [
720,
1280,
3
],
"names": [
"height",
"width",
"channel"
],
"info": {
"video.height": 720,
"video.width": 1280,
"video.codec": "h264",
"video.pix_fmt": "yuv420p",
"video.is_depth_map": false,
"video.fps": 15,
"video.channels": 3,
"has_audio": false
}
}
}
}
```

## Getting started

```py
# pip install lerobot
from lerobot.datasets.lerobot_dataset import LeRobotDataset

dataset = LeRobotDataset("jnogga/droid_failure")
```

See [droid_example.ipynb](droid_example.ipynb) for a more detailed example.

## Citation

All credit goes to the original authors of DROID. If you find their work helpful, please cite

**BibTeX:**

```bibtex
@article{khazatsky2024droid,
title = {DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset},
author = {Alexander Khazatsky and Karl Pertsch and Suraj Nair and Ashwin Balakrishna and Sudeep Dasari and Siddharth Karamcheti and Soroush Nasiriany and Mohan Kumar Srirama and Lawrence Yunliang Chen and Kirsty Ellis and Peter David Fagan and Joey Hejna and Masha Itkina and Marion Lepert and Yecheng Jason Ma and Patrick Tree Miller and Jimmy Wu and Suneel Belkhale and Shivin Dass and Huy Ha and Arhan Jain and Abraham Lee and Youngwoon Lee and Marius Memmel and Sungjae Park and Ilija Radosavovic and Kaiyuan Wang and Albert Zhan and Kevin Black and Cheng Chi and Kyle Beltran Hatch and Shan Lin and Jingpei Lu and Jean Mercat and Abdul Rehman and Pannag R Sanketi and Archit Sharma and Cody Simpson and Quan Vuong and Homer Rich Walke and Blake Wulfe and Ted Xiao and Jonathan Heewon Yang and Arefeh Yavary and Tony Z. Zhao and Christopher Agia and Rohan Baijal and Mateo Guaman Castro and Daphne Chen and Qiuyu Chen and Trinity Chung and Jaimyn Drake and Ethan Paul Foster and Jensen Gao and Vitor Guizilini and David Antonio Herrera and Minho Heo and Kyle Hsu and Jiaheng Hu and Muhammad Zubair Irshad and Donovon Jackson and Charlotte Le and Yunshuang Li and Kevin Lin and Roy Lin and Zehan Ma and Abhiram Maddukuri and Suvir Mirchandani and Daniel Morton and Tony Nguyen and Abigail O'Neill and Rosario Scalise and Derick Seale and Victor Son and Stephen Tian and Emi Tran and Andrew E. Wang and Yilin Wu and Annie Xie and Jingyun Yang and Patrick Yin and Yunchu Zhang and Osbert Bastani and Glen Berseth and Jeannette Bohg and Ken Goldberg and Abhinav Gupta and Abhishek Gupta and Dinesh Jayaraman and Joseph J Lim and Jitendra Malik and Roberto Martín-Martín and Subramanian Ramamoorthy and Dorsa Sadigh and Shuran Song and Jiajun Wu and Michael C. Yip and Yuke Zhu and Thomas Kollar and Sergey Levine and Chelsea Finn},
year = {2024},
}
```

license: cc-by-4.0
任务类别：
- 机器人学
标签：
- LeRobot
- 机器人操作
展示名称：DROID 失败任务集
规模类别：
- 1M<n<10M
---

## DROID 失败任务集

[DROID-COMMUNITY](https://droid-dataset.github.io/) 中的失败任务集。已剔除数据不一致的任务集，最终保留了全部失败任务集的约91%。

本数据集从原始1.0.1版本的全分辨率数据转换为LeRobotDataset v3.0格式（占用空间0.25 TiB，包含1.5k个索引节点），并补充了来自[KarlP/droid](https://huggingface.co/KarlP/droid/tree/main)的额外标注。

<div align="center" style="margin: 16px 0;">
<video controls autoplay loop muted playsinline style="max-width: 100%; border-radius: 10px;">
<source src="https://huggingface.co/datasets/jnogga/droid_failure/resolve/main/droid_example_episode.mp4" type="video/mp4">
您的浏览器不支持视频标签。
</video>
</div>

如需查看成功任务集，请访问[jnogga/droid_success](https://huggingface.co/datasets/jnogga/droid_success)。

## 更新记录

* 2026年2月22日：移除了摄像头分辨率异常的任务集。

## 数据集结构

根据任务集的不同，外置摄像头被分配为左视图或右视图。关于其外参，已提供来源以及对应的评估指标与评分。

[meta/info.json](meta/info.json):
json
{
"codebase_version": "v3.0",
"fps": 15,
"data_files_size_in_mb": 100.0,
"video_files_size_in_mb": 200.0,
"robot_type": "franka_emika_panda_robotiq_2F-85",
"total_episodes": 13747,
"total_frames": 3476037,
"total_tasks": 13747,
"chunks_size": 1000,
"splits": {
"train": "0:13747"
},
"data_path": "data/chunk-{chunk_index:03d}/file_{file_index:03d}.parquet",
"video_path": "videos/{video_key}/chunk-{chunk_index:03d}/file_{file_index:03d}.mp4",
"features": {
"action.cartesian_position": {
"dtype": "float32",
"shape": [7],
"names": ["position.x", "position.y", "position.z", "quaternion.w", "quaternion.x", "quaternion.y", "quaternion.z"],
"fps": 15
},
"action.cartesian_velocity": {
"dtype": "float32",
"shape": [7],
"names": ["position.x", "position.y", "position.z", "quaternion.w", "quaternion.x", "quaternion.y", "quaternion.z"],
"fps": 15
},
"action.gripper_position": {
"dtype": "float32",
"shape": [1],
"names": null,
"fps": 15
},
"action.gripper_velocity": {
"dtype": "float32",
"shape": [1],
"names": null,
"fps": 15
},
"action.joint_position": {
"dtype": "float32",
"shape": [7],
"names": ["joint_0", "joint_1", "joint_2", "joint_3", "joint_4", "joint_5", "joint_6"],
"fps": 15
},
"action.joint_velocity": {
"dtype": "float32",
"shape": [7],
"names": ["joint_0", "joint_1", "joint_2", "joint_3", "joint_4", "joint_5", "joint_6"],
"fps": 15
},
"observation.cartesian_position": {
"dtype": "float32",
"shape": [7],
"names": ["position.x", "position.y", "position.z", "quaternion.w", "quaternion.x", "quaternion.y", "quaternion.z"],
"fps": 15
},
"observation.gripper_position": {
"dtype": "float32",
"shape": [1],
"names": null,
"fps": 15
},
"observation.joint_position": {
"dtype": "float32",
"shape": [7],
"names": ["joint_0", "joint_1", "joint_2", "joint_3", "joint_4", "joint_5", "joint_6"],
"fps": 15
},
"observation.joint_velocity": {
"dtype": "float32",
"shape": [7],
"names": ["joint_0", "joint_1", "joint_2", "joint_3", "joint_4", "joint_5", "joint_6"],
"fps": 15
},
"wrist_camera_intrinsics": {
"dtype": "float32",
"shape": [4],
"names": ["fx", "cx", "fy", "cy"],
"fps": 15
},
"wrist_camera_to_robot_extrinsics": {
"dtype": "float32",
"shape": [7],
"names": ["position.x", "position.y", "position.z", "quaternion.w", "quaternion.x", "quaternion.y", "quaternion.z"],
"fps": 15
},
"left_external_camera_intrinsics": {
"dtype": "float32",
"shape": [4],
"names": ["fx", "cx", "fy", "cy"],
"fps": 15
},
"left_external_camera_to_robot_extrinsics": {
"dtype": "float32",
"shape": [7],
"names": ["position.x", "position.y", "position.z", "quaternion.w", "quaternion.x", "quaternion.y", "quaternion.z"],
"fps": 15
},
"left_external_camera_to_robot_source": {
"dtype": "string",
"shape": [1],
"names": null,
"fps": 15
},
"left_external_camera_to_robot_metric": {
"dtype": "string",
"shape": [1],
"names": null,
"fps": 15
},
"left_external_camera_to_robot_score": {
"dtype": "float32",
"shape": [1],
"names": null,
"fps": 15
},
"left_external_camera_to_robot_extrinsics_raw": {
"dtype": "float32",
"shape": [7],
"names": ["position.x", "position.y", "position.z", "quaternion.w", "quaternion.x", "quaternion.y", "quaternion.z"],
"fps": 15
},
"right_external_camera_intrinsics": {
"dtype": "float32",
"shape": [4],
"names": ["fx", "cx", "fy", "cy"],
"fps": 15
},
"right_external_camera_to_robot_extrinsics": {
"dtype": "float32",
"shape": [7],
"names": ["position.x", "position.y", "position.z", "quaternion.w", "quaternion.x", "quaternion.y", "quaternion.z"],
"fps": 15
},
"right_external_camera_to_robot_source": {
"dtype": "string",
"shape": [1],
"names": null,
"fps": 15
},
"right_external_camera_to_robot_metric": {
"dtype": "string",
"shape": [1],
"names": null,
"fps": 15
},
"right_external_camera_to_robot_score": {
"dtype": "float32",
"shape": [1],
"names": null,
"fps": 15
},
"right_external_camera_to_robot_extrinsics_raw": {
"dtype": "float32",
"shape": [7],
"names": ["position.x", "position.y", "position.z", "quaternion.w", "quaternion.x", "quaternion.y", "quaternion.z"],
"fps": 15
},
"keep_interval_index": {
"dtype": "int64",
"shape": [1],
"names": null,
"fps": 15
},
"frame_index": {
"dtype": "int64",
"shape": [1],
"names": null,
"fps": 15
},
"timestamp": {
"dtype": "float32",
"shape": [1],
"names": null,
"fps": 15
},
"index": {
"dtype": "int64",
"shape": [1],
"names": null,
"fps": 15
},
"task_index": {
"dtype": "int64",
"shape": [1],
"names": null,
"fps": 15
},
"episode_index": {
"dtype": "int64",
"shape": [1],
"names": null,
"fps": 15
},
"observation.images.left_external": {
"dtype": "video",
"shape": [720, 1280, 3],
"names": ["height", "width", "channel"],
"info": {
"video.height": 720,
"video.width": 1280,
"video.codec": "h264",
"video.pix_fmt": "yuv420p",
"video.is_depth_map": false,
"video.fps": 15,
"video.channels": 3,
"has_audio": false
}
},
"observation.images.wrist": {
"dtype": "video",
"shape": [720, 1280, 3],
"names": ["height", "width", "channel"],
"info": {
"video.height": 720,
"video.width": 1280,
"video.codec": "h264",
"video.pix_fmt": "yuv420p",
"video.is_depth_map": false,
"video.fps": 15,
"video.channels": 3,
"has_audio": false
}
},
"observation.images.right_external": {
"dtype": "video",
"shape": [720, 1280, 3],
"names": ["height", "width", "channel"],
"info": {
"video.height": 720,
"video.width": 1280,
"video.codec": "h264",
"video.pix_fmt": "yuv420p",
"video.is_depth_map": false,
"video.fps": 15,
"video.channels": 3,
"has_audio": false
}
}
}
}

## 快速入门

py
# 安装依赖：pip install lerobot
from lerobot.datasets.lerobot_dataset import LeRobotDataset

dataset = LeRobotDataset("jnogga/droid_failure")

如需更详细的使用示例，请参阅[droid_example.ipynb](droid_example.ipynb)。

## 引用声明

本数据集的所有成果均归功于DROID的原始作者。若您的工作使用了该数据集，请引用如下BibTeX格式的文献：

**BibTeX:**

bibtex
@article{khazatsky2024droid,
title = {DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset},
author = {Alexander Khazatsky and Karl Pertsch and Suraj Nair and Ashwin Balakrishna and Sudeep Dasari and Siddharth Karamcheti and Soroush Nasiriany and Mohan Kumar Srirama and Lawrence Yunliang Chen and Kirsty Ellis and Peter David Fagan and Joey Hejna and Masha Itkina and Marion Lepert and Yecheng Jason Ma and Patrick Tree Miller and Jimmy Wu and Suneel Belkhale and Shivin Dass and Huy Ha and Arhan Jain and Abraham Lee and Youngwoon Lee and Marius Memmel and Sungjae Park and Ilija Radosavovic and Kaiyuan Wang and Albert Zhan and Kevin Black and Cheng Chi and Kyle Beltran Hatch and Shan Lin and Jingpei Lu and Jean Mercat and Abdul Rehman and Pannag R Sanketi and Archit Sharma and Cody Simpson and Quan Vuong and Homer Rich Walke and Blake Wulfe and Ted Xiao and Jonathan Heewon Yang and Arefeh Yavary and Tony Z. Zhao and Christopher Agia and Rohan Baijal and Mateo Guaman Castro and Daphne Chen and Qiuyu Chen and Trinity Chung and Jaimyn Drake and Ethan Paul Foster and Jensen Gao and Vitor Guizilini and David Antonio Herrera and Minho Heo and Kyle Hsu and Jiaheng Hu and Muhammad Zubair Irshad and Donovon Jackson and Charlotte Le and Yunshuang Li and Kevin Lin and Roy Lin and Zehan Ma and Abhiram Maddukuri and Suvir Mirchandani and Daniel Morton and Tony Nguyen and Abigail O'Neill and Rosario Scalise and Derick Seale and Victor Son and Stephen Tian and Emi Tran and Andrew E. Wang and Yilin Wu and Annie Xie and Jingyun Yang and Patrick Yin and Yunchu Zhang and Osbert Bastani and Glen Berseth and Jeannette Bohg and Ken Goldberg and Abhinav Gupta and Abhishek Gupta and Dinesh Jayaraman and Joseph J Lim and Jitendra Malik and Roberto Martín-Martín and Subramanian Ramamoorthy and Dorsa Sadigh and Shuran Song and Jiajun Wu and Michael C. Yip and Yuke Zhu and Thomas Kollar and Sergey Levine and Chelsea Finn},
year = {2024},
}