on 28 July 2023
The first IEEE&OpenAtomCARSMOS Intelligent Mobility Open Source Algorithm International Challenge, hosted by the Ministry of Industry and Information Technology of the People’s Republic of China, is jointly organized by the Open Atom Open Source Foundation (OpenAtom Foundation), IEEE (International Institute of Electrical and Electronics Engineers), and CARSMOS Yuanao, a subsidiary of the Open Atom Open Source Foundation. It is co-organized by the project team and CARLA, an internationally renowned open-source simulation community for autonomous driving. The technical supporters of the competition include Synkrotron Technologies, Futurewei Technologies, Shanghai AI Lab, Tsinghua MarsLab, OpenEV, etc. The CARSMOS Challenge is open to the world, and excellent teams and players from domestic and foreign universities, scientific research institutes, open source organizations, and industry organizations can sign up for the competition. The challenge includes the preliminary round (online competition part) and the rematch (offline competition part), and provides the CARSMOS open source code warehouse ( https://github.com/carsmos ) as a benchmark reference design and implementation (non-mandatory conditions). The part of the online competition is based on the simulation environment, providing multiple pre-defined training scenarios for the participating teams to learn and debug, and then use more complex test scenarios to evaluate and compete each participating team’s autonomous driving algorithm in highways and urban areas. Perception, planning and decision-making capabilities in complex traffic environments. The winning teams from the online competition will enter the offline competition to determine the final winner.
The preliminary round of the challenge is an online competition in the simulation system, based on the Oasis autonomous driving simulation platform donated by SenseTech to the CARSMOS open source intelligent travel project. The main goal of the competition is to evaluate the driving ability of autonomous driving algorithms in realistic traffic scenarios. The scoring standard is to obey the traffic rules, deal with various road conditions, and arrive at the destination safely within the specified time. During the competition, each participating team has multiple opportunities to upload the Docker file package that implements the automatic driving algorithm and submit it to the background scoring system of the competition to automatically run and get a score, and then continue to optimize, and then submit to get a higher score. The final score shall prevail.
The final part of the CARSMOS Challenge is an offline competition and awards event, which is planned to be held in the city where the 2023 Open Atom Open Source Competition is held. Each participating team loads its own open-source algorithm model on the fully debugged competition vehicle provided by the competition committee, and safely completes the specified competition on the road in the actual test area, and the referee group scores according to multiple dimensions. The awarding ceremony of the CARSMOS Challenge will be held at the awarding ceremony of the 2023 Open Source Competition together with other track challenges of the Foundation.
1) Algorithm Open Source – CARSMOS Challenge is the only open source challenge among the global autopilot competitions, which means that participating teams are required to open source the algorithm within 3 months, which will help future generations to continue to improve and make the competition better and better powerful.
2) The event is of high standard – it is jointly organized by IEEE, an internationally renowned professional academic institution, and the Ministry of Industry and Information Technology, which is in charge of China’s auto industry.
3) High degree of internationalization – guided by IEEE; attracting international teams.
4) Low barriers to entry – CARSMOS has provided a very modular benchmark design, which is very convenient for participating teams to replace any of the operator modules or subsystems to test their own innovative algorithms.
5) Large talent reserve – CARSMOS Challenge aims to attract and train college students/graduate students to enter the field of autonomous driving and cultivate a large number of reserve talents. The winners are outstanding engineers and future leaders that leading companies are competing to recruit.
Carsmos open source smart travel project group (please refer to the English official website https://www.carsmos.ai/Chinese official website https://www.carsmos.cn/ ) is a strategic open source project group under the OpenAtom Foundation , led by Shenxin Kechuang, Huawei America Research Institute (FutureWei), Beijixiongxin, Open Motors, currently has Changan Automobile, China Automobile Intelligent Control, Tsinghua Interdisciplinary Information Research Institute and other enterprises and institutions to participate, and is still in the process of continuous development and growth middle. The project has a technical steering committee composed of world-renowned experts, including Li Keqiang, academician of the Chinese Academy of Engineering and professor of Tsinghua University; Guan Xiaohong, academician of the Chinese Academy of Sciences and Xi’an Jiaotong University; Academician, Professor Moshe Vardi of Rice University and others. The Yuanao CARSMOS project group has 6 open source software and hardware sub-projects, namely the open source application of autonomous driving and the community JUNO open source project, the open source project of DORA, the middleware of autonomous driving and robot operating system, and the vehicle-road collaborative edge computing service and management platform ESSA Open source projects, Oasis open source project for autonomous driving simulation test platform, EDITEV open source project for intelligent electric vehicle design, and Hermes open source project for ChipletD2D interconnection interface standard for NPU computing chips in the automotive field. Yuanao CARSMOS open source smart travel project group is committed to building a unified and open smart car hardware, computing, software and ecological platform, enabling smart driving open source basic software and hardware framework solutions, creating industry standards, enriching industrial ecology, cultivating industrial talents, and promoting Industry-university-research cooperation progresses.
The mission of Yuanao Project Group includes:
1)The world’s first automotive open source project community, builds an open source ecosystem, gathers leading technologies, and enters the era of data-driven development 2.0 from the era of logic-driven development 1.0;
2)To develop and achieve China’s contribution in the software ecology of the smart car industry;
3)Cultivate talents in algorithms, software and hardware that are synchronized with the development of the automobile industry;
4)Accelerate the industrialization from innovative technology to commercial application.
1.Competition registration： https://competition.atomgit.com/competitionInfo?id=2e1cce10c89711edb4b22fd906d12a1e
2.Download the automatic driving simulation system Oasis： https://docs.carsmos.cn/#/zh-cn/release?id=_912-%e4%b8%8b%e8%bd%bd
3.Familiar with the automatic driving simulation system Oasis, and develop your own algorithm by referring to the document： https://docs.carsmos.cn/#/
4.Refer to the document submission algorithm, and wait for the running result after the submission is successful. The registration for the competition has been opened, and it is currently the stage of the practice competition. You can participate in the competition now, get familiar with it first, and have a chance to get better results and win more generous rewards! Please scan the QR code to join the WeChat group of CARSMOS Challenge Service:
If the group has exceeded the number limit, please add the WeChat account (ID: synkrotron1) of SYNKROTRON Assistant, and remark “Contest” to join.