Skip to content

robodrive-24/toolkit

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

155 Commits
docs
docs
 
 
track-1
track-1
 
 
track-2
track-2
 
 
track-3
track-3
 
 
track-4
track-4
 
 
track-5
track-5
 
 
README.md
README.md
 
 

Repository files navigation

RoboDrive Toolkit

Twitter website slack wechat ICRA2024

Welcome to The RoboDrive Challenge! 👋

🚘 About

RoboDrive is one of the first competitions that targeted probing the Out-of-Distribution (OoD) robustness of state-of-the-art autonomous driving perception models, centered around two mainstream topics: common corruptions and sensor failures.

There are eighteen real-world corruption types in total, ranging from three perspectives:

  • Weather and lighting conditions, such as bright, low-light, foggy, and snowy conditions.
  • Movement and acquisition failures, such as potential blurs caused by vehicle motions.
  • Data processing issues, such as noises and quantizations happen due to hardware malfunctions.

Additionally, we aim to probe the 3D scene perception robustness under camera and LiDAR sensor failures:

  • Loss of certain camera frames during the driving system sensing process.
  • Loss of one or more camera views during the driving system sensing process.
  • Loss of the roof-top LiDAR view during the driving system sensing process.

Kindly visit our webpage to explore more details and instructions for this challenge. 🚙

🌈 Venue

This competition is affiliated with the 41st IEEE Conference on Robotics and Automation (ICRA 2024).


ICRA is the IEEE Robotics and Automation Society's flagship conference. ICRA 2024 will be held from May 13th to 17th, 2024, in Yokohama, Japan.

🏆 Winning Team

We are glad to announce the winning teams of the 2024 RoboDrive Challenge.

  • Track 1: Robust BEV Detection
    • 🥇 DeepVision, 🥈 Ponyville Autonauts Ltd, 🥉 CyberBEV
  • Track 2: Robust Map Segmentation
    • 🥇 SafeDrive-SSR, 🥈 CrazyFriday, 🥉 Samsung Research
  • Track 3: Robust Occupancy Prediction
    • 🥇 ViewFormer, 🥈 APEC Blue, 🥉 hm.unilab
  • Track 4: Robust Depth Estimation
    • 🥇 HIT-AIIA, 🥈 BUAA-Trans, 🥉 CUSTZS
  • Track 5: Robust Multi-Modal BEV Detection
    • 🥇 safedrive-promax, 🥈 Ponyville Autonauts Ltd, 🥉 HITSZrobodrive

📚 Technical Report

Competition Report

"The RoboDrive Challenge: Drive Anytime Anywhere in Any Condition", Technical Report, 2024.

@article{kong2024robodrive_challenge,
    title = {The RoboDrive Challenge: Drive Anytime Anywhere in Any Condition},
    author = {Lingdong Kong and Shaoyuan Xie and Hanjiang Hu and Yaru Niu and Wei Tsang Ooi and Benoit R. Cottereau and Lai Xing Ng and Yuexin Ma and Wenwei Zhang and Liang Pan and Kai Chen and Ziwei Liu and Weichao Qiu and Wei Zhang and Xu Cao and Hao Lu and Ying-Cong Chen and Caixin Kang and Xinning Zhou and Chengyang Ying and Wentao Shang and Xingxing Wei and Yinpeng Dong and Bo Yang and Shengyin Jiang and Zeliang Ma and Dengyi Ji and Haiwen Li and Xingliang Huang and Yu Tian and Genghua Kou and Fan Jia and Yingfei Liu and Tiancai Wang and Ying Li and Xiaoshuai Hao and Yifan Yang and Hui Zhang and Mengchuan Wei and Yi Zhou and Haimei Zhao and Jing Zhang and Jinke Li and Xiao He and Xiaoqiang Cheng and Bingyang Zhang and Lirong Zhao and Dianlei Ding and Fangsheng Liu and Yixiang Yan and Hongming Wang and Nanfei Ye and Lun Luo and Yubo Tian and Yiwei Zuo and Zhe Cao and Yi Ren and Yunfan Li and Wenjie Liu and Xun Wu and Yifan Mao and Ming Li and Jian Liu and Jiayang Liu and Zihan Qin and Cunxi Chu and Jialei Xu and Wenbo Zhao and Junjun Jiang and Xianming Liu and Ziyan Wang and Chiwei Li and Shilong Li and Chendong Yuan and Songyue Yang and Wentao Liu and Peng Chen and Bin Zhou and Yubo Wang and Chi Zhang and Jianhang Sun and Hai Chen and Xiao Yang and Lizhong Wang and Dongyi Fu and Yongchun Lin and Huitong Yang and Haoang Li and Yadan Luo and Xianjing Cheng and Yong Xu},
    journal = {arXiv preprint arXiv:2405.08816}, 
    year = {2024},
}

Track Reports

Track 1: Robust BEV Detection

Track 2: Robust Map Segmentation

Track 3: Robust Occupancy Prediction

Track 4: Robust Depth Estimation

Track 5: Robust Multi-Modal BEV Detection

Outline

ℹ️ Useful Info

# Item Link
🌐 Competition Webpage https://robodrive-24.github.io
🔧 Competition Toolkit https://github.com/robodrive-24/toolkit
:octocat: Official GitHub Account https://github.com/robodrive-24
📫 Contact [email protected]

🕐 Timeline

Note: All timestamps are in the AoE (Anywhere on Earth) format.

  • Dec 25 '23 - Team Up; Register for your team by filling in this Google Form.
  • Jan 05 '24 - Release of training and evaluation data.
  • Jan 15 '24 - Competition servers online @ CodaLab.
  • Mar 31 '24 - Phase One deadline.
  • Apr 30 '24 - Phase Two deadline.
  • May 17 '24 - Award decision announcement @ ICRA 2024.

🚙 Challenge Tracks

There are five tracks in this RoboDrive challenge, with emphasis on the following 3D scene perception tasks:

# Task Description Doc Server
Track 1 Robust BEV Detection Evaluating the resilience of detection algorithms against diverse environmental and sensor-based corruptions [Link] [Link]
Track 2 Robust Map Segmentation Focusing on the segmentation of complex driving scene elements in BEV maps under varied driving conditions [Link] [Link]
Track 3 Robust Occupancy Prediction Testing the accuracy of occupancy grid predictions in dynamic and unpredictable real-world driving environments [Link] [Link]
Track 4 Robust Depth Estimation Assessing the depth estimation robustness from multiple perspectives for comprehensive 3D scene perception [Link] [Link]
Track 5 Robust Multi-Modal BEV Detection Tailored for evaluating the reliability of advanced driving perception systems equipped with multiple types of sensors [Link] [Link]

Our evaluation servers were developed based on the CodaLab platform. 🙆‍♂️


CodaLab is an open-source web-based platform that enables researchers, developers, and data scientists to collaborate, with the goal of advancing research fields where machine learning and advanced computation are used.

♨️ Data Preparation

Kindly refer to DATA_PREPARE.md for the details to prepare the training and evaluation data.

🚀 Getting Started

Kindly refer to GET_STARTED.md to learn more usage of this toolkit.

📝 Changelog

  • Mar 28 '24 - Phase 2 evaluation data of Tracks 1 to 5 have been released.
  • Jan 22 '24 - The evaluation server of Track 4 is online. [Server4]
  • Jan 15 '24 - The evaluation server of Track 5 is online. [Server5]
  • Jan 15 '24 - The evaluation server of Track 3 is online. [Server3]
  • Jan 15 '24 - The evaluation server of Track 2 is online. [Server2]
  • Jan 15 '24 - The evaluation server of Track 1 is online. [Server1]
  • Jan 12 '24 - Instructions, baseline models, and results of Track 5 have been released.
  • Jan 12 '24 - Training and evaluation data of Track 5 have been released.
  • Jan 08 '24 - A list of Frequently Asked Questions (FAQs) has been summarized for better clarity.
  • Jan 06 '24 - Instructions, baseline models, and results of Tracks 1 to 3 have been released.
  • Jan 05 '24 - Training and evaluation data of Tracks 1 to 4 have been released.
  • Dec 25 '23 - Register for your team by filling in this Google Form.
  • Dec 01 '23 - Launch of The RoboDrive Challenge at ICRA 2024. More info coming soon!

🥇 Awards

The top-performing participants of this competition are honored with cash awards and certificates.

Award Amount Honor
🥇 1st Place Cash Award $ 5000 Official Certificate
🥈 2nd Place Cash Award $ 3000 Official Certificate
🥉 3rd Place Cash Award $ 2000 Official Certificate

Note: The cash awards are donated by our sponsors and are shared among five tracks. We reserve the right to examine the validity of each submission. For more information, kindly refer to the Terms & Conditions section.

Additionally, we provide the following awards for participants that meet certain conditions.

Award Honor Condition
Innovative Award Official Certificate Solutions with excellent novelty
Certificate of Participation Official Certificate Teams with submission records in both phases

License

Creative Commons License
This competition is under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

❤️ Sponsor

This competition is supported by HUAWEI Noah's Ark Lab.


The Noah’s Ark Lab is the AI research center for Huawei Technologies, aiming to make significant contributions to both the company and society by innovating in artificial intelligence, data mining, and related fields.

Citation

If you find this competition helpful for your research, kindly consider citing our papers:

@article{xie2023robobev,
    title = {RoboBEV: Towards Robust Bird's Eye View Perception under Corruptions},
    author = {Shaoyuan Xie and Lingdong Kong and Wenwei Zhang and Jiawei Ren and Liang Pan and Kai Chen and Ziwei Liu},
    journal = {arXiv preprint arXiv:2304.06719}, 
    year = {2023}
}
@inproceedings{kong2023robodepth,
    title = {RoboDepth: Robust Out-of-Distribution Depth Estimation under Corruptions},
    author = {Lingdong Kong and Shaoyuan Xie and Hanjiang Hu and Lai Xing Ng and Benoit R. Cottereau and Wei Tsang Ooi},
    booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
    year = {2023},
}
@inproceedings{kong2023robo3d,
    title = {Robo3D: Towards Robust and Reliable 3D Perception against Corruptions},
    author = {Lingdong Kong and Youquan Liu and Xin Li and Runnan Chen and Wenwei Zhang and Jiawei Ren and Liang Pan and Kai Chen and Ziwei Liu},
    booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
    pages = {19994-20006},
    year = {2023},
}
@misc{mmdet3d,
    title = {MMDetection3D: OpenMMLab Next-Generation Platform for General 3D Object Detection},
    author = {MMDetection3D Contributors},
    howpublished = {\url{https://github.com/open-mmlab/mmdetection3d}},
    year = {2020}
}

In the meantime, kindly cite the technical reports of the nuScenes dataset and the CodaLab platform:

@inproceedings{caesar2020nuscenes,
    title={nuScenes: A Multimodal Dataset for Autonomous Driving},
    author={Holger Caesar and Varun Bankiti and Alex H. Lang and Sourabh Vora and Venice Erin Liong and Qiang Xu and Anush Krishnan and Yu Pan and Giancarlo Baldan and Oscar Beijbom},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    pages = {11621-11631},
    year={2020}
}
@article{pavao2023codalab,
    title = {CodaLab Competitions: An Open Source Platform to Organize Scientific Challenges},
    author = {Adrien Pavão and Isabelle Guyon and Anne-Catherine Letournel and Dinh-Tuan Tran and Xavier Baró and Hugo Jair Escalante and Sergio Escalera and Tyler Thomas and Zhen Xu},
    journal = {Journal of Machine Learning Research (JMLR)},
    pages = {1-6},
    year = {2023}
}

⚖️ Terms & Conditions

This competition is made freely available to academic and non-academic entities for non-commercial purposes such as academic research, teaching, scientific publications, or personal experimentation. Permission is granted to use the data given that you agree:

  1. That the data in this competition comes “AS IS”, without express or implied warranty. Although every effort has been made to ensure accuracy, we do not accept any responsibility for errors or omissions.
  2. That you may not use the data in this competition or any derivative work for commercial purposes such as, for example, licensing or selling the data, or using the data with a purpose of procuring a commercial gain.
  3. That you include a reference to RoboDrive (including the benchmark data and the specially generated data for academic challenges) in any work that makes use of the benchmark. For research papers, please cite our preferred publications as listed on our webpage.

To ensure a fair comparison among all participants, we require:

  1. All participants must follow the exact same data configuration when training and evaluating their algorithms. Please do not use any public or private datasets other than those specified for model training.
  2. The theme of this competition is to probe the out-of-distribution robustness of autonomous driving perception models. Therefore, any use of the corruption and sensor failure types designed in this benchmark is strictly prohibited, including any atomic operation that comprises any one of the mentioned corruptions.
  3. To ensure the above two rules are followed, each participant is requested to submit the code with reproducible results before the final result is announced; the code is for examination purposes only and we will manually verify the training and evaluation of each participant's model.

🤔 Frequently Asked Questions
🤔 Q1: "How can I register a valid team for this competition?"
🚙 A1: To register a team, kindly fill in this Google Form. The registration period is from now till the deadline of phase one, i.e., Mar 31 '24.
🤔 Q2: "Are there any restrictions for the registration? For example, the number of team members."
🚙 A2: Each team leader should make a valid registration for his/her team. Each participant can only be registered by one team. There is no restriction on the number of team members in a team.
🤔 Q3: "Whether team members can be changed during the competition?"
🚙 A3: No. You CANNOT change the list of team members after the registration. You must register again as a new team if you need to add or remove any members of your team.
🤔 Q4: "How many tracks can I participate in?"
🚙 A4: Each team can participate in at most two tracks in this competition.
🤔 Q5: "What can I expect from this competition?"
🚙 A5: We provide the winning teams from each track with cash awards 💰 and certificates 🥇. The winning solutions will be summarized as a technical report 📖. An example of last year's technical report can be found here.
🤔 Q6: “Can I use additional data resources for model training?"
🚙 A6: No. All participants must follow the SAME data preparation procedures as listed in DATA_PREPARE.md. Additional data sources are NOT allowed in this competition.
🤔 Q7: "Can I use corruption augmentations during model training?"
🚙 A7: For Track 1-4: No. The theme of this competition is to probe the out-of-distribution robustness of autonomous driving perception models. Therefore, all participants must REFRAIN from using any corruption simulations as data augmentations during the model training, including any atomic operation that comprises any one of the corruptions in this competition. For Track 5, there is no limitation on the augmentations.
🤔 Q8: "How should I configurate the model training? Are there any restrictions on model size, image size, loss function, optimizer, number of epochs, and so on?"
🚙 A8: We provide one baseline model for each track in GET_STARTED.md. The participants are recommended to refer to these baselines as the starting point in configuring the model training. There is no restriction on normal model training configurations, including model size, image size, loss function, optimizer, and number of epochs.
🤔 Q9: "Can I use LiDAR data for Tracks 1 to 4?"
🚙 A9: Only RAW LiDAR points data is allowed for Tracks 1 to 4 in training (e.g., generate sparse depth map). During inference, Tracks 1 to 4 are single-modality tracks that only involve the use of camera data. The goal of these tracks is to probe the robustness of perception models under camera-related corruptions. Participants who are interested in multi-modal robustness (camera + LiDAR) can refer to Track 5 in this competition.
🤔 Q10: "Is it permissible to use self-supervised model pre-training (such as MoCo and MAE)?"
🚙 A10: Yes. The use of self-supervised pre-trained models is possible. Such models may include MoCo, MoCo v2, MAE, DINO, and many others. Please make sure to acknowledge (in your code and report) if you use any pre-trained models.
🤔 Q11: "Can I use large models (such as SAM) to generate pre-training or auxiliary annotations?"
🚙 A11: No. The use of large foundation models, such as CLIP, SAM, SEEM, and any other similar models, is NOT allowed in this competition. This is to ensure a relatively fair comparing environment among different teams. Any violations of this rule will be regarded as cheating and the results will be canceled.
🤔 Q12: "Are there any restrictions on the use of pre-trained weights (such as DD3D, ImageNet, COCO, ADE20K, Object365, and so on)?"
🚙 A12: Following the most recent BEV perception works, it is possible to use pre-trained weights on DD3D, ImageNet, and COCO. The use of weights pre-trained on other datasets is NOT allowed in this competition.
🤔 Q13: "Can I combine the training and validation sets for model training?"
🚙 A13: It is strictly NOT allowed to use the validation data for model training. All participants MUST follow the nuScenes official train split during model training and REFRAIN from involving any samples from the validation set. Any violations of this rule will be regarded as cheating and the results will be canceled.
🤔 Q14: "Can I use model ensembling and test-time augmentation (TTA)?"
🚙 A14: Like many other academic competitions, it is possible to use model ensembling and test-time augmentation (TTA) to enhance the model when preparing the submissions. The participants SHOULD include necessary details for the use of model ensembling and TTA in their code and reports.
🤔 Q15: "How many times can I make submissions to the server?"
🚙 A15: For phase one (Jan. - Mar.), a team can submit up to 3 times per day and 99 times total. For phase two (Apr.), a team can submit up to 2 times per day and 49 times total. One team is affiliated with one CodaLab account only. Please REFRAIN from having multiple accounts for the same team.
🤔 Q16: " Can I use pretrained denoising or deblurring models during inference?"
🚙 A16: No. The goal of the competition is to develop a more robust perception model and using pre-trained denoising models is out of the scope of this competition.
🤔 Q17: " Can I use augmentation other than the corruption methods used in the competition?"
🚙 A17: Similar to Q7, you can use data augmentation methods that do NOT include the corruption simulation algorithms used in the competition. More details of the used corruptions can be found from this technical report.
🤔 Q18: " What is the sensor corruptions in Track-5?"
🚙 A18: For the camera sensor, we use camera corruptions by setting all the pixels to 0. For the LiDAR sensor, we use random points drop, drop points within certain view field angles, and beam drop.
🤔 Q19: " What is the depth estimation metric for Track 4?"
🚙 A19: We use RELATIVE depth estimation, not absolute depth estimation for evaluation.
🤔 ...
🚙 ...

Contact

📫 Didn't find a related FAQ to your questions? Let us know ([email protected])!

Organizers

Affiliation

🙆‍♂️ Acknowledgements

This competition is developed based on the RoboBEV, RoboDepth, and Robo3D projects.

This competition toolkit is developed based on the MMDetection3D codebase.


MMDetection3D is an open-source toolbox based on PyTorch, towards the next-generation platform for general 3D perception. It is a part of the OpenMMLab project developed by MMLab.

The evaluation sets of this competition are constructed based on the nuScenes dataset from Motional AD LLC.

Part of the content of this toolkit is adopted from The RoboDepth Challenge @ ICRA 2023.

About

Official Toolkit for The RoboDrive Challenge

robodrive-24.github.io

Resources

Readme
Activity

Stars

73 stars

Watchers

5 watching

Forks

9 forks
Report repository

Releases

No releases published

Packages

No packages published

Contributors 3

Languages