The RoboMaster is a platform for robotic competitions and academic exchange founded
by Da-Jiang Innovations (DJI) and designed specifically for global
technology enthusiasts.
Since its establishment in 2013, RoboMaster has been committed to its mission -
"empowering young learners to transform their world with the power of engineering and technology",
in order to attract bright engineering talents with interdisciplinary abilities and showcase the possibilities of technological
innovation to audiences around the globe.
The Institute for AI Industry Research (AIR), Tsinghua University,
was founded in 2020 by Dr. Ya-Qin Zhang, a world-class scientist and entrepreneur in multi-media and AI. It focuses on applied research geared
towards the internationalization, intelligentization, and industrialization of the fourth technological revolution. Driven by
academia-industry innovations simultaneously, AIR aims at making breakthroughs in core AI technologies, training future industry leaders, and
achieving industrial leapfrogging progress.
AIR, Tsinghua University, in collaboration with the RoboMaster Organizing Committee, will launch the 2nd RoboMaster University
Sim2Real Challenge (RMUS) at ICRA 2023. In this challenge, the organizer will provide a
downloadable simulator and a cloud-based robotic game field.
Throughout the process of completing a specific rearrangement task, participants can develop navigation and grasping algorithms with
the simulator and can remotely access the physical robot in the game field for a limited number of verifications.
Software Architecture of RMUS
Over the past five years, RoboMaster challenges have been held uninterruptedly at ICRA and have attracted more than 100 teams worldwide from universities and robotics labs. Last year, 117 teams from MIT, CMU, ETH, ICL, TUM, NUS, Tsinghua, and other universities participated in the first RMUS. For ICRA 2023, we propose our new RMUS with an additional adversarial mechanism and look forward to further expanding our impact.
  Date | Activities | Notes |
---|---|---|
  Nov 14, 2022 | Release of Competition Rules and Relevant Materials | RoboMaster official website |
  Nov 14, 2022 - Dec 31, 2022 | Formal Registration | RoboMaster official website |
  Nov 14, 2022 - Mar 1, 2023 | Simulator Test | Teams submit codes on the official website. |
  Dec 1, 2022 - Mar 1, 2023 | Sim2Real Test | The committee downloads the team code. Real-site testing and feedback data will be provided. |
  Mar 1, 2023 - Mar 15, 2023 | Mineral Search and Exchange Task Ranking Competition Technical Assessment Review |
Based on the ranking of the Sim2Real Test, 1. Qualified teams are shortlisted. 2. Double-confirming the competition |
  Mar 16, 2023 - May 15, 2023 | AI Adversarial Test | The committee downloads the team code. Real-site testing and feedback data will be provided. |
  May 20, 2023 | Code Submission | Each team is allowed to submit the codes only once. |
  ICRA 2023 | Formal Match | Participating teams are not required to arrive on site. |
Notice: The time of the Formal Match will be based on the local time zone of the competition site, while the time of other matches is based on UTC + 8 (Beijing time).
The RoboMaster University Sim2Real Challenge (RMUS) is an international competition open to all university students. As long as you are passionate about robotics and eager to compete, it's time to register for RMUS, which will be held at ICRA 2023.
Registration system: https://registration.robomaster.com/registration-center?eventType=team&id=235
If you have any questions, you can contact us via:
Email: robomaster@air.tsinghua.edu.cn
Telegram: https://t.me/+UBU60LQfxppmMThl
WeChat ID: AIR_DISCOVER_Lab
The Simulator Test lasts from November 14, 2022 to March 1, 2023. During this stage, teams will focus on developing algorithms in a simulated environment and completing the mineral search and exchange task according to the Rules Manual.
After the local simulator test is completed, the image can be submitted multiple times on the specified test platform (please see GitHub instructions for details) to complete the online simulation test. The teams that complete three ore grabbing within five minutes in online simulator test can enter the Sim2Real test.
If you encounter any problems in the Simulator Test, you can submit the problems to the issue on GitHub, find answers in the FAQ List (English) or FAQ List (Chinese) (updated every Monday), or contact us via email, Telegram, or WeChat.
Simulator Test Instrucion (GitHub): https://github.com/AIR-DISCOVER/ICRA-RM-Sim2real-Baseline
Simulator Test Instrucion (Gitee): https://gitee.com/tb5zhh/icra-rm-sim2real-baseline
The Sim2Real Test runs from 1 December, 2022 to 1 March, 2023. In this stage, the main goal for competing teams is to adjust and optimize the algorithm for the mineral search and exchange task using test data from the real site.
Newest Notifacation: Chinese English
A technical report (Chinese templete) (English templete) is required by March 1, 2023, at 23:59 (UTC + 8). Finalists (those teams able to pass the Sim2Real Test) will be determined based on the overall score:
70% Sim2Real Highest Test score + 30% Technical Report score
The finalists will be announced here. If you have any questions, please contact us via email, Telegram, or WeChat.
Each university or college may have more than one team participating in the competition. However, only the team from each university or college with the highest overall score will advance to the next stage. Each college or university might merge its participating teams. Before the Formal Match, the final team members must be determined through the registration system.
Notice: The duration of the weekly field test is limited. The schedule will be posted on the official page for submissions. Teams should plan their schedules accordingly. The submission deadline for team codes is March 1.
The list of teams that enter the AI Adversairial Test stage is as follows. If you have any questions, please feel free to contact us via email.
University | Team | Test Score | Report Score | Total Score |
---|---|---|---|---|
Northwestern Polytechnical University | Firefly | 66.00 | 24.50 | 90.50 |
Harbin Institute of Technology, Weihai | SOT | 64.88 | 23.50 | 88.38 |
Xi’an Jiaotong University | Gold Miners | 57.52 | 24.00 | 81.52 |
Beijing Institute of Technology | Dream Chaser | 56.87 | 22.60 | 79.47 |
Tsinghua University | GLGG | 56.16 | 22.00 | 78.16 |
Harbin Institute of Technology, Shenzhen | Critical HIT | 46.51 | 23.75 | 70.26 |
South China Agricultural University | Dream Pioneer | 48.43 | 21.50 | 69.93 |
Northeastern University | TDT | 44.64 | 24.85 | 69.49 |
Dongguan University Of Technology | Nazarite | 21.00 | 22.50 | 43.50 |
Shenzhen Technology University | SZTU | 21.00 | 22.50 | 43.50 |
China institute of atomic energy | FTM | 21.00 | 22.25 | 43.25 |
South China Normal University | Prettycool | 21.00 | 21.75 | 42.75 |
The AI Adversarial Test runs from March 16, 2023 to May 15, 2023. The teams will be selected for the AI Adversarial Test based on their ranking in the Sim2Real Test. Each team will compete for two months against the AI algorithm provided by the committee in order to fine-tune their strategy for mineral distribution adjustment task & mineral search and exchange task. Each team is permitted to test their algorithms in simulation with no limitations and has the opportunity to sim2real test once per week.
The Formal Match will be held in London, United Kingdom (date and time to be confirmed).
The final competition consists of elimination rounds. In this stage, each team will take on the role of artificial intelligence to set up random minerals for the opposing team in one round and vice versa in the other. The score is calculated on the number of minerals exchanged. If the scores are identical, the winner is the participant who completed the mineral search and exchange task in the shortest amount of time. The teams will be ranked and then given certificates and bonuses.
The winning teams are listed below. If you have any questions, please feel free to contact us via email.
  Prize | Ranking | Team Name | Team Chinese Name |   University |
---|---|---|---|---|
Grand Price | First Place | FTM | FTM | China Institute of Atomic Energy |
  First Prize | Second Place | Dream Pioneer | 梦想开拓者 | South China Agricultural University |
Third Place | GLGG | GLGG说的全都对 | Tsinghua University | |
Fourth Place | Gold Miners | 黄金矿工 | Xi’an Jiaotong University | |
Second Prize | TDT | TDT | Northeastern University | |
Firely | Firely | Northwestern Polytechnical University | ||
Dream Chaser | 追梦 | Beijing Institute of Technology | ||
Third Prize | Criticial Hit | 南工骁鹰 | Harbin Institute of Technology, Shenzhen | |
SZTU | SZTU | Shenzhen Technology University | ||
SOT | SOT | Harbin Institute of Technology, Weihai | ||
Pretty Cool | 炫酷 | South China Normal University | ||
Nazarite | 行者 | Dongguan University Of Technology | ||
Participation Award | DUXING YOUTH | 笃行青年 | Xi’an Jiaotong University | |
Duxing | 笃行 | Xi’an Jiaotong University |
At its core, RMUS enables participants to gain points by remotely programming a customized RoboMaster EP to run autonomously on an official game field. Each game lasts five minutes, during which robots rearrange minerals in accordance with the information displayed on the exchange tags in order to gain points. Participants will be ranked according to their total points gained.
Rules Manual (English): RoboMaster 2023 University Sim2Real Challenge Rules Manual v1.0
Rules Manual (Chinese): RoboMaster 2023 高校Sim2Real挑战赛比赛规则手册v1.0
  Prize | Ranking | Quantity |     Awards |
---|---|---|---|
  Grand Price | First Place | 1 |
|
  First Prize | Second Place | 1 |
|
Third Place | 1 |
|
|
Fourth Place | 1 |
|
|
  Second Prize | Fifth-Eighth Place | 4 |
|
  Third Prize | The teams that enter the AI Adversarial Test but do not make it to the quarterfinals | (Several) To be determined based on the results of all teams |
|
  Participation Award | The teams that enter the Sim2Real Test but do not enter the AI Adversarial Test | (Several) To be determined based on the results of all teams |
|
Guyue Zhou, Associate Professor, AIR, Tsinghua
University.
Lu Fang, Associate Professor, Dept. of EE, Tsinghua
University.
Xianyuan Zhan, Assistant Professor, AIR, Tsinghua
University.
Jiangtao Gong, Assistant Professor, AIR, Tsinghua
University.
Hao Zhao, Research Scientist, Intel.
Tao Kong, Senior Researcher, ByteDance AI Lab.
Jie Song, Research Scientist, ETH.
Lu Shi, University of California, Riverside.
Zhanxiang Cao, Tsinghua University.
Beiwen Tian, Tsinghua University.
Yufei Jia, Tsinghua University.
Zehao Li, University of New South Wales.
robomaster@air.tsinghua.edu.cn
Telegram
(010) 82150160
Copyright @ 2023 AIR Sim2Real Committee & RoboMaster Organizing Committee. All Rights Reserved