Header Search Widget
Robot Open Autonomous Racing (ROAR™)

Simulation Racing Series

ROAR provides a Python-based racing simulation environment that allows contenders to race their autonomous AI agents. Contenders can fully train and test their AI algorithms without any vehicle hardware in this series. 

Code Submissions

Fall 2024 Submissions (Open)

Final submissions will be due on Friday, December 13. If you are having any issues submitting your agent, please email (hc_kuan@berkeley.edu). For Fall 2024, we are using the Monza Map v1.1.

The official Berkeley Monza Map v1.1 is provided here. 

Award Criteria:

Important: By participating in ROAR Competition and submitting your code, you give the ROAR Committee the permission to evaluate and publicly release your competition results on our website.

Contestants who meet the following criteria will be awarded one or more of ROAR Competition Awards:

  • Grand Prize ($500 Amazon gift card): 
    This is awarded to the team who achieves the fastest total time after the completion of three (3) laps. Any major collision events will result in the competition server to reset the starting point at Lap 0. Minor collisions that do not trigger reset will not further cause any penalty to the final score.
  • Second-Place Prize ($200 Amazon gift card):
    This is awarded to the team who achieves the second fastest total time after the completion of three (3) laps. Any major collision events will result in the competition server to reset the starting point at Lap 0. Minor collisions that do not trigger reset will not further cause any penalty to the final score.
  • Third-Place Prize ($100 Amazon gift card):
    This is awarded to the team who achieves the third fastest total lap time after the completion of three (3) laps. Any major collision events will result in the competition server to reset the starting point at Lap 0. Minor collisions that do not trigger reset will not further cause any penalty to the final score.
  • Prime-Directive Award:
    This is awarded to the team who achieves the fastest total time after the completion of three (3) laps and without encountering any major or minor collision during the competition.

Summer 2024 Schedule

Voluntary Test Script Submissions are due on Friday, December 6, 11:59 (PST) Midnight

  • Competitors can submit a test script to Huo Chao Kuan (hc_kuan@berkeley.edu) by providing your GitHub repository.

  • Competitors can adjust their script as necessary before the Final Competition.

Mandatory Final Submissions are due on Friday, December 13 (PST) Midnight

Instructions

Starting from Fall 2023, ROAR Competition will be running on Monza Map with the ROAR_PY repository as the reference code base. Please see Installation Documentation here.

  1. Set up ROAR_PY
    1. Clone ROAR_PY repo to desired location on your local machine by running:
      >>> git clone https://github.com/augcog/ROAR_PY.git
      (You should not make any changes to this repo)
    2. You might need to set up the ssh key on your local machines for git. The tutorial can be found here: Adding a new SSH key to your GitHub account – GitHub Docs
    3. Following the guide on README in the ROAR_PY folder to install dependencies. The documentation is also linked here: Introduction – ROAR_PY Documentation (gitbook.io). You will need to complete “Interface Installation” and “Implementation Installation” sections to finish the setup. At this point, you should create a conda environment locally. The later instructions should be run under this environment. 
  2. Set up Competition Script
    1.  Go to ROAR_Competition repo at FHL-VIVE-Center-for-Enhanced-Reality/ROAR_Competition: Internal Collaboration Repo (github.com), And create your own fork by clicking on the “Fork” button.
    2. Clone the competition script to desired location on your local machine by running:
      >>> git clone path_to_your_repo

3. Set up the map

    1. Download the Monza map from official website
    2. Since the map is made on Carla 0.9.12, you will need to upgrade/downgrade your Carla package to this specific version by running the following command in your conda environment:
      >>> pip install carla==0.9.12
    3. Double click on the .exe file to start the map.

4. Code and run your solution

    1. Go to ROAR_PY_Competition/competition_code, you will see three files above
    2. Your solution should be coded in submission.py. You can add additional files if needed, but you should not make any changes to competition_runner.py and infrastructure.py. An example solution can be found under the testing branch.
    3. If you have any additional packages or special instructions for your solution, please include them in a file called instruction.txt and add it to your solution repo.
    4. Push your solution to github and submit your repo link as your submission by form below.
Submission Form for 2024 Fall Simulation Series
  • Please include first and last name
  • Make sure competition_runner.py is there.
  • If your GitHub is not open-sourced, please provide instructions for download. By default, we will not publicly disclose your submission code without your explicit permission. Nevertheless, we strongly encourage contestants to open source their solutions via the Berkeley ROAR platform.
  • Ex: High School Senior, College 2nd year, etc.