Skip to content

Racing working group

Jump to bottom
Rahul Mangharam edited this page Oct 26, 2023 · 12 revisions

The Racing Working Group was established to lead the development of the integration of Autoware.Universe to autonomous racing environments. This working group is open to anyone that is related to autonomous racing and provides knowledge in this field, including those who are not members of the Autoware Foundation.

Working group leaders

Working group goals

The goals of the Racing Working Group are:

  1. Bring the Autoware.Universe autonomous driving software stack to autonomous race vehicles like the F1TENTH racecar, Indy Autonomous Challenge (IAC) racecar, and Autonomous Gokart.
  2. Create Simulation environments (race vehicles, race tracks) that enable virtual testing of autonomous racecars.
  3. Develop new algorithms for the Autoware.Universe software that operate an autonomous vehicle at at high speeds in racetrack environments. This includes the following:
    • Object detections for speeds up to 300 km/h
    • Visual localization methods for localize the vehicle on racetracks with little landmarks for speeds up to 300 km/h
    • Multi Vehicle path planning
  4. Develop new algorithms for the Autoware.Universe software that operate an autonomous vehicle at the limits of handling at high speeds in adversarial scenarios. This includes the following:
    • Raceline generation (Minimum curvature, minimum time)
    • Behavior planner that reacts to race rules
    • High dynamic path planner for overtaking maneuvers hat high speeds and high accelerations
    • Multi vehicle Interaction and aggressive driving
    • Balancing safety and performance
  5. Evaluate Autoware.Universe on high performance autonomous vehicles that operate on the limits of the vehicle dynamics at high speeds in adversarial scenarios.
  6. Create new small-scale hardware platforms (1:10 scale and 1:3 scale) that enable testing of Autoware.Universe and its functionalities faster than on classical passenger cars.
  7. Create University training programs that will teach ROS2 and Autoware.Universe in the context of autonomous racing. These programs are open source and will provided on the openEDX platform F1TENTH Courses.
  8. Provide support for international hackathons and racing events where small-scale vehicles race against each other.

Current activities

  • Hardware-F1Tenth: The Racing Working Group is currently working on the development of a newer version of the 1:10 scale vehicles. This includes an update from the Jetson Xavier computing devices to more flexible platforms and a resign of the mechatronic and sensing layout.
  • Hardware-Gokart: The Racing Working Group is currently working on the development of a standardized 1:3 scale EV Autonomous gokart. This will complement the F1Tenth as a larger scale vehicle to support the testing of more complex modules and algorithms.
  • Software-F1Tenth: The Racing Working Group is currently working on the development of the Autoware.Universe integration on the F1TENTH vehicle. This includes the development of all relevant ROS2 drivers for the F1TENTH vehicle (perception, localization, planning, control), the development of the Autoware.Universe interface for F1TENTh and the testing of Autoware.Universe on the vehicle.
  • Software-Gokart: The Racing Working Group is currently working on the development of the Autoware.Universe integration on the autonomous gokart. This includes the development of all relevant ROS2 drivers for the gokart (perception, localization, planning, control), the development of a containerized environment, and the testing of Autoware.Universe on the vehicle.
  • Software-IAC: The Racing Working Group is currently working on the development of the Autoware.Universe, particularly the perception module integration on the Indy Autonomous Challenge race car. Next competition will be held summer 2023 in Mozan, Italy.

Ongoing Competitions

Results and Achievements:

  • F1TENTH vehicle as Standard Platform: We provide the Bill of Material and an in-depth build manual (slides, videos) to set up a small-scale autonomous racing vehicle. This vehicle is the standard testing platform for Autoware.Auto, is allowed to race in racing events and is the baseline for further developments.

  • Racing ODD Definition: This file includes a high level definition of what we need to take care about when it comes to autonomous racing with Autoware.Universe . We define the possible physical infrastructures, operation limits, possible objects to detect etc. This is compliant to other Autoware.Universe ODDs and creates a machine readable file for the automatic generation of test scenarios.

  • Racing in SVL Simulator: The SVL simulator enables the simulation of racing vehicles (F1TENTH, IAC racecar) as well es different racetrack environments (Indianapolis Motor Speedway, Indianapolis Motor Speedway 1:10 scale, DiY Robocar Circuit Launch)

  • EV Grand Prix: The first autonomous gokart competition held at Purdue University, West Lafayette with participants from top universities such as University of California San Diego, Georgia Tech, Purdue University, etc.

  • Talks and Workshop: The topic of autonomous racing is increasing its popularity over time and different researchers organized workshops on this particular topic. You can see what is going in the community here:

Meeting minutes

See the discussions with the Meeting:racing_wg label

Get Started on Autonomous Racing

Autonomous Racing is a special application in the field of autonomous driving. Different researchers around the world developed algorithms for driving autonomously at the dynamic limits. To get started in the field of autonomous racing we are recommending the following literature and code repositories:

  • Autonomous Vehicles on the Edge: A Survey on Autonomous Vehicle Racing. This paper is the first-ever holistic survey on all the research that has happened in the field of autonomous racing. It provides a clear and structured overview on all the efforts in the field of software and hardware development. Look into this paper first to see what happened already and get and idea of the state of the art. In addition, a list of the papers can be found in this Github Repository

  • Learn to Drive (and Race!) Autonomous Vehicles?. This education lecture was made for the Embedded Systems Week 2021 and gives a short introduction about Autonomous Driving and Racing. Code is provided for three different algorithms that will race an F1TENTH car in the F1TENTH gym.

  • What we can learn from autonomous racing?. This paper gives a high-level overview why autonomous motorsport is interesting for R&D. This explains both the challenges and the opportunities we have by solving the problem of autonomous racing.

  • Optimal Raceline Generation. A racecar needs to follow an optimal raceline to be as fast as possible. This repository includes an optimization algorithm for a minim curvature and minimum time optimization.

  • Racetrack Database: To run an autonomous vehicle on a racetrack you need to have data about the racetrack layout. We provide the racetrack layouts (2D representation) of international racetracks for normal scaled vehicle and 1:10 vehicles:

How to participate

The Racing Working Group is meeting every Thursday virtually where you can join via Zoom. Have a look in the Autoware Calender to get the times of the meeting.

Clone this wiki locally