What we do
The 2021 AI Driving Olympics Compete in the 2021 edition of the Artificial Intelligence Driving Olympics (AI-DO 6)! The AI-DO serves to benchmark the state
Compete in the 5th AI Driving Olympics (AI-DO) The 5th edition of the Artificial Intelligence Driving Olympics (AI-DO 5) has officially started! The AI-DO serves
Duckietown and NVIDIA work together for accessible AI and robotics education: Meet the NVIDIA powered Duckiebot
Duckietown and NVIDIA partnership for accessible AI and robotics education NVIDIA GTC, October 6, 2020: Duckietown and NVIDIA align efforts to push the boundaries of
The AI Driving Olympics (AI-DO) is back! We are excited to announce the launch of the AI-DO 3, which will culminate in a live competition event
Team JetBrains came out on top on all 3 challenges It was a busy (and squeaky) few days at the International Conference on Robotics and
RT @PAVECampaign: There is still time to register for PAVE’s next virtual panel “A Look Inside PAVE’s Annual Gift Guide” – tomorrow at 2pm…
The AI Driving Olympics #AIDO2021 is now underway! Participate in Urban Driving, Advanced Perception, and Racing leagues and reach the finals at #NeurIPS2021!
@motionaldrive @SwissRe @awscloud
#Robotics #AutonomousVehicles #AI #competition https://t.co/EDTrOxwkc2
Neural networks are at the basis of solving advanced perception challenges, and allow to shift focus from building great models to collecting large datasets.
#learningautonomy #robotics #stem #AI
How do Self-driving cars give meaning to what they see? Advanced visual perception encompasses many tasks, e.g., object detection or semantic segmentation.
#learningautonomy #robotics #stem #AI
Cameras produce a lot of data! But how do machines make sense of it? One approach is extracting “features” of interest for the task at hand. In this video, we talk about how to filter images and extract useful information from camera data.
The information embedded in camera pixel measurements often needs to be “decoded” for robots to make use of it. The key to this are the camera intrinsic and extrinsic matrices, which can be obtained through a camera calibration procedure. How does it work?
Compact, affordable, and able to measure at a distance: these are some of the traits that make cameras ubiquitous in robotics. But how do 2D images relate to the 3D world they measure? In this video, we look at what projective geometry is.
Proportional, Integrative, Derivative (PID) control is the most successful control approach of all times.
Learn more about the Duckietown massive online open course “Self-Driving Cars with #Duckietown” on
#Robots can estimate where they are in the world using odometry.
In this video from the “Self Driving Cars with #Duckietown” MOOC on @edXOnline, we’ll see how to use wheel encoders and a motion model to derive and deploy an algorithm on #Duckiebots.
Mathematical models allow us to predict the future. But are they always trustworthy?
In this video from the “#selfdrivingcars with #Duckietown” MOOC on @edXOnline we discuss the modeling of a differential drive #robot.
Get started with the platform
The Duckietown project
Duckietown started as a class at MIT in 2016. You can watch the “duckumentary” created about the first class.
Duckietown is now a worldwide initiative to realize a new vision for AI/robotics education.
Since 2018 the project is coordinated by the non-profit Duckietown Foundation.
The Duckietown platform
The platform has two parts: Duckiebots and Duckietowns.
Duckiebots are low-cost mobile robots that are built almost entirely from off-the-shelf parts. The only onboard sensor is the forward-facing camera.
Duckietowns are the roads, which are constructed from exercise mats and tape, and the signage which the robots use to navigate around.
Duckietown for education
The Duckietown platform designed as part of an a university AI/robotics curriculum.
It has been used in prestigious universities, such as MIT, ETH Zürich, and Université de Montréal.
We are developing a “class-in-a-box” that comprises lectures, exercises, and theory, that combine with the physical robot platform to reinforce the core concepts.
If you are an instructor interested in using Duckietown, read here to get started.
Duckietown for research
The Duckietown platform has also been used extensively for research on mobile robotics and physically embodied AI systems.
If you are a researcher, read more about getting started about using the platform for research. See also: papers using Duckietown, the researchers using Duckietown.
Duckietown for “Makademics”
Makademics (makers + academics) are people who want to learn and build on their own and also want a very deep understanding of how things are working.
We want to allow everybody to learn AI/robotics even if they are not at elite institutions like MIT and ETH Zürich.
With Duckietown you can easily build your own robot, and follow along our lectures and interact with a global community of learners.