Self-Driving Car Challenge

Brought to you by

Udacity Didi
Join Now

WIN $100,000 and a chance to run your code on Udacity's self-driving car!


  • Person Icon

    $100,000 winning prize

  • Person Icon

    5 top teams will fly to Silicon Valley for the awards ceremony

  • Person Icon

    5 teams will run their code on Udacity's Self-Driving Car

About the Challenge

Udacity is building an open source self-driving car , and we’re challenging you to contribute! Join this competition and help us build the world's first open source autonomous vehicle !

We are launching a first-of-its kind competition with Didi Chuxing. The focus will be on a core feature of self-driving cars—the Automated Safety and Awareness Processing Stack (ASAPS), which identifies stationary and moving objects from a moving car, using data that includes Velodyne point cloud, radar objects, and camera image frames.

Your challenge? Create a redundant, safe, and reliable system for detecting hazards, that allows for pedestrian, vehicle, and general obstacle detection useful to both human drivers and self-driving car systems.

Anyone from anywhere around the world is invited to participate. The challenge has two rounds, the first of which begins on March 22, 2017. The winning team will receive $100,000 in cash, and have the opportunity to see their code in action on Udacity’s Self-Driving car.

Winners will be subject to the terms and Official Rules .


Participants will need to have an understanding of the following frameworks and skills

  • Ubuntu Linux & Python, C++
  • Machine Learning Classification techniques
  • Robot Operating System (ROS)
  • Euclidean Clustering Algorithms (DBSCAN, Mean Shift, etc.)
  • Point Cloud Library (PCL)
  • Ground plane segmentation and removal
  • Computer Vision, Image Processing, Object Recognition
  • Temporal tracking and sensor-to-sensor registration


This challenge is all about transformative technologies and forward-thinking skills—exactly what these Nanodegree programs teach. You may already be enrolled in one or more, in which case you’re not only an ideal challenge competitor, you’re part of a future-minded vanguard that’s going to change the world! If you haven’t enrolled yet, do so today. This challenge wants you, and the future needs you.

Our Partner

Didi Chuxing is the world's largest mobile tech-based transportation platform

Car App

Didi Chuxing offers a full range of personal and corporate mobility services from its app to 400 million users, using machine-learning, ride-matching, and dispatching algorithms. We call for innovators and disruptors around the world to join us in helping cities solve transportation, environmental, and employment challenges by harnessing the power of AI and big-data technologies.

About Didi


  • How do I get started?

    Just follow these simple steps!

    1. Click "Join Now" to start the process.

      • If you are already signed in to a Udacity account, you’ll be taken to a Registration page. Some of your information will auto-populate; you’ll need to fill in the rest.
      • If you are NOT already signed in, then clicking “Join Now” will take you to a Udacity account page. (Note: you don’t need to be a Udacity student to participate, but you do need to have have a Udacity account to participate.)
    2. Select "Sign In" or "Sign Up".

      • Select "Sign In" if you already have a Udacity account. When you sign in, you’ll be taken to a Registration page. Fill in whatever doesn’t auto-populate. Completing this step will take you to the first Challenge page.
      • Select "Sign Up" if you need to create a Udacity account. This will take you to a Registration page. Provide ALL the requested details to create your account, including the information required for your Udacity account, and the information required for your challenge registration. Note: this will be a two-page process. When you’ve completed these steps, you’ll then be taken to the first Challenge page.
    3. At the Challenge page, you’ll be able to create your own team, or accept an invitation to join another team. Note: You must be part of a team—even if it’s a team of one!—to participate. You have three options for team participation:

      • You can start your own team, and then invite others by email to join it.
      • You can accept an invitation to join another team.
      • You can visit the Forum to connect with other participants, and get a team established that way.
    4. Once your team is created, you’re ready for the release of the first dataset, which will happen on March 22, 2017!
  • How many members can be on a team?

    There is no limit to team size, but only participants registered pursuant to the terms detailed in our Contest Rules will be considered official team members. In addition, only a maximum of two members will be flown to Silicon Valley, should your team be a top five team.

  • Who owns the rights to the code that we write?

    All code will be open sourced and made available on Udacity sites as part of the Udacity Open Source Self-Driving Car Project. See Contest Rules for full details.

  • How will you prevent teams from using automated submissions bots to guess the solution and hack the leaderboard?

    Teams must submit their code to progress from Round 1 to Round 2, and code for all finalists will be reviewed.

  • When do the rounds start?

    Round 1 will start on March 22, 2017 and Round 2 will start on May 1, 2017.

  • Will all contestants get a chance to run our code on the Udacity Self-Driving Car?

    The top five teams that come to the final award ceremony will have a chance to run their contest submitted code on the Udacity Self-Driving Car.

    See Contest Rules for certain restrictions and prize eligibility requirements.

Icon globe

Udacity 现已提供中文版本! A Udacity tem uma página em português para você! There's a local version of Udacity for you! Sprechen Sie Deutsch?

Besuchen Sie und entdecken Sie lokale Angebote, unsere Partnerunternehmen und Udacitys deutschsprachigen Blog.

前往优达学城中文网站 Ir para a página brasileira Go to Indian Site Icon flag de Zu continue in English