Duckietown Profile Learning Experiences User Guide
- June 17, 2024
- DUCKIETOWN
Table of Contents
- Duckietown Profile Learning Experiences
- Specifications:
- Product Information:
- Duckietown Profile:
- The Duckietown Project:
- Our Community and Team:
- The Duckietown Ecosystem:
- Getting Started:
- Duckietown for Research:
- Q: What age group is Duckietown suitable for?
- Q: Can Duckietown be used for commercial applications?
Duckietown Profile Learning Experiences
“`html
Specifications:
-
Product Name: Duckietown
-
Description: State-of-the-art robotics and AI ecosystem for
learning, teaching, and research -
Features: Hardware, software, cutting-edge tools,
workflows -
Learning Experiences: Accessible, engaging, fun
Product Information:
Duckietown is a state-of-the-art robotics and AI ecosystem
designed to facilitate learning, teaching, and research. It brings
together hardware, software, cutting-edge tools, and workflows to
demystify the science and technology of modern robot autonomy. The
goal is to provide tangible learning experiences that are
accessible, engaging, and fun, nurturing talents to shape the
future of autonomy.
Duckietown Profile:
The Duckietown project aims to address the challenges posed by
the robotics and AI revolution. It provides a versatile, open, and
engaging ecosystem with state-of-the-art content, tools, and
workflows to equip individuals with real-world skills and
competence in the field of robotics and AI.
The Duckietown Project:
Originating as a class at MIT in 2016, Duckietown has evolved
into a worldwide initiative for AI and robotics education. It is
utilized in classrooms and labs of prestigious universities
globally, including MIT, ETH, Imperial College, Tsinghua
University, and Oxford University.
Our Community and Team:
The Duckietown community comprises members from over 62
countries, including 175 universities and 115 companies/labs. The
team includes experts with Ph.D. qualifications in robotics and
AI.
The Duckietown Ecosystem:
The hardware component of Duckietown includes Duckiebots,
minimal autonomy platforms designed for self-driving cars. Equipped
with cameras, encoders, IMU, Raspberry Pi, Jetson Nano, DC motors,
LEDs, and ample memory and power supply, Duckiebots are
top-of-the-line in their class.
Product Usage Instructions:
Getting Started:
-
Unbox your Duckietown kit and familiarize yourself with the
components. -
Charge the Duckiebot using the provided power supply.
-
Install any required software on your computer or device as per
the instructions provided. -
Connect the Duckiebot to your computer or device using the
necessary cables.
Duckietown for Research:
If you are using Duckietown for research purposes, ensure to
follow the guidelines provided by your institution or project
supervisor. Utilize the Duckiebot’s sensing capabilities and
computational power to conduct experiments and gather data for
analysis.
FAQ:
Q: What age group is Duckietown suitable for?
A: Duckietown is designed for learners of all ages who are
interested in robotics and AI education. While it is commonly used
in university settings, younger students can also benefit from its
engaging learning experiences with proper supervision.
Q: Can Duckietown be used for commercial applications?
A: While Duckietown primarily focuses on education and research,
some components of the ecosystem may be adapted for commercial
applications. It is recommended to consult with the Duckietown team
for specific commercial use cases.
“`
Learning Autonomy
DUCKIETOWN
Official Information Guide
1 Duckietown: Learning Autonomy ————————————-
3 The Duckietown Project —————————————————-
5 Our Community and Team ————————————————-
7 The Duckietown Ecosystem ———————————————–
Example Curriculum
11 ———————————————————
12 Getting Started ———————————————————————–
13 Duckietown for Research —————————————————
15 Testimonials ————————————————————————–
DuckietownLearning Autonomy
Duckietown Profile
Duckietown is the state-of-the-art robotics and AI ecosystem for learning,
teaching and doing research. Bringing together hardware, software, cutting-
edge tools and workflows, Duckietown demystifies and democratizes the science
and technology of modern robot autonomy. Duckietown provides tangible learning
experiences that are accessible, engaging and fun, nurturing talents in our
generation to shape the next generations fo autonomy.
” AI and robotics are the most beautiful disciplines it’s mankind’s attempt
at creating artificial creatures that think and act like us.
AI and robotics will change our world. Everybody should understand the
possibilities, the current status and how much is left to do.
1
”
BEAUTY FUN
IMPORTANCE CHALLENGES
2
The Duckietown Project
The Robotics and AI revolution
The fourth industrial revolution has started: robotics and AI technologies are
becoming increasingly part of our daily lives.
Existing educational systems are inadequate at producing enough qualified
workforce or informed citizenry to support this revolution.
Duckietown provides an ecosystem that is versatile, open, engaging, and
designed with state-of-the-art content, tools and workflows to transmit real-
world skills and competence.
Revenue (USD M)
40000
AI market worldwide
30000
20000
10000
0 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
AI at Duckietown in action
3
The Duckietown project
The history of Duckietown
Duckietown started as a class at the Massachusetts Institute of Technology in
2016, for the students of the Computer Science and Artifcial Intelligence
Laboratory (CSAIL) and is now a worldwide initiative to realize a new vision
for AI and robotics education.
Duckietown is used in the classrooms and laboratories of some of the world’ s
best universities: MIT, ETH, Imperial College, Tsinghua University, Oxford
University, etc.
4
OUR COMMUNITY AND TEAM
Jacopo Tani,Ph. D.
Prof. Liam Paull, Ph. D
Andrea Censi, Ph. D.
Andrea Francesco Dan.iele
Prof. Emilio Frazzoli, Ph. D.
SLACK MEMBERS
2100
UNIVERSITIES
175
COMPANIES & LABS
115
COUNTRIES
62
The Duckietown Ecosystem
THE HARDWARE
Duckiebot
Duckiebots are minimal autonomy platforms: the simplest self-driving cars with
which we can do real science. Designed through the years to meet the needs of
teachers, professionals and researchers worldwide, Duckiebots are the top of
line in their class.
Sensing: Camera, Encoders, IMU, Time of Flight Computation: Raspberry Pi,
Jetson Nano Actuation: DC motors, LEDs Memory: 64GB, class 10 Power: 5V, 10Ah
Duckietown
Duckietowns are structured and modular environments built on two layers: road
and signal, to offer a repeatable but flexible driving experience, without
fixed maps.
Modularity: assemble and combine fundamental building blocks Structure:
appearance specifications (colors, geometries) guaran-
tee functionality Smart City: road and signal layers can be augmented with a
network of smart traffic lights and watchtowers to create a real city-robot.
Duckiedrone
Duckiedrones are fully-programmable DIY quadcopters designed to introduce
younger audiences to automation science and technology through exciting flying
machines.
Sensing: IR, Camera, IMU Computation: Raspberry Pi Actuation: DC motors, LEDs
Memory: 32GB, class 10
7
LEARNING EXPERIENCE
The Duckietown Ecosystem
Theory modules
Software solutions
Hardware deployment
Our 3 pilars of software development Modularity
Containerization technology allows adding, mix and matching with extreme ease
State of the art tools
Learning the tools used in industry is as important as the art of it.
Open source content
“From a box of parts, to a robotic ecosystem, in only 1452 steps, without
hiding anything”.
Run a program wherever you want
The Duckietown software architecture uses ROS, Docker and Python to enable
versatile workflows: in simulation (locally or on the cloud) as well as on
Duckiebots in Ducketown Autolabs, or on Duckiedrones!
8
The Duckietown Ecosystem
THE DUCKIETOWN ECOSYSTEM
The Duckietown ecosystem is more than a sum of robots. It is a collection of
tools and services seamlessly integrated with our hardware environment.
Duckietown bridges learners with instructors, online with offline, simulation
with hardware, talents with institutions.
Community of instructors and learners
9
The Duckietown Ecosystem
WHO IS DUCKIETOWN FOR?
No matter how expert you are, Duckietown will help on your jou-r ney from
discovery to mastery, for “Duckietown is a place of joy and relaxed
introspection.”
Learners can join offline classes with their instructors or learn in autonomy
with Duckietown’s massive open online course on edX.
Instructors benefit from Duckietown’s class-in-a-box: the one-click solution
for learning autonomy, inclusive of curricula, hardware, auto-grading and
support for troubleshooting.
Researchersgain access to a programmable infrastructure for reproducible devel
opment and benchmarking of robotic agents, and an international competition
featured at ICRA and NeurIPS.
10
Exemplary Curriculum
EXAMPLE CURRICULUM
Operation tools
introduction to the basic tools
Construction Configuration Operation Calibration
Autonomy
Exploring real-world solutions to the challenges of robot autonomy -from
theory to deployment
Perception Control Planning Behaviours
Machine Learning
Exploring AI methods and workflows for solving challenging autonomous tasks.
Supervised Learning Reinforcement Learning
System Development
set up an efficient software environment for robotics with state of the art
ROS Git Docker Test/data driven development Open source dynamics
11
Getting started
Getting started has never been easier
Learners: join the “Self-Driving Cars with Duckietown” MOOC Researchers: join
the AI Driving Olympics Instructors: get a class-in-a-box
MASSIVE ONLINE OPEN COURSE
Self-Driving cars with Ducki etown is the world’s first robot autonomy MOOC
with hardware!
AI DRIVING OLYMPICS
The AI-DO are an international scientific competition established in 2018 to
bench mark the state of the art of machine learning approaches to solving
self-driving cars technology challenges. AI-DO finals have been held at ICRA
and NuerIPS.
200+
challenges
135000+
evaluation jobs
12
Duckietown for research
DUCKIETOWN FOR RESEARCH
The DUCKIENet
The Decentralized Urban Collaborative Benchmarking Environment Network
(DUCKIENet) provides an accessible and reproducible framework focused on
autonomous vehicle fleets operating in model urban environments. The DUCKIENet
enables users to develop and test a wide variety of different algorithms and
then deploy them locally in simulation, locally on a robot, in a cloud-based
simulation, or on a real robot in a remote lab. In each case, the submitter
receives feedback and scores based on well-defined metrics.
The Duckietown Automated Laboratories
One of the central components of the DUCKIEnet is the Duckietown Autolab
(DTA), a remotely accessible standardized setup that is itself also relatively
low-cost and reproducible.
DTAs include an off-the-shelf camera-based localization system. The
accessibility of the hardware testing environment through enables experimental
benchmarking that can be performed on a network of DTAs in different
geographical locations.
13
Duckietown for research
The system is validated by analyzing the repeatability of experiments
conducted using the infrastructure and show that there is low variance across
different robot hardware and across different remote labs.
Looking forward
In Duckietown we have the contention that there is a need for stronger efforts
towards reproducible research for robotics, and that to achieve this we need
to consider the evaluation in equal terms as the algorithms themselves. In
this fashion, we can obtain reproducibility by design through the research and
development processes.
Reproducible benchmarking in robotics
As robotics matures and increases in complexity, it is more necessary than
ever that robot autonomy research be reproducible. Duckietown offers a new
concept for reproducible robotics research that integrates development and
benchmarking from the beginning of the research/development processes.
14
Testimonials
VOICE OF OUR MEMBERS
“Duckietown was much more than just a class, it was a hands-on deep dive into
hardware, software, and systems integration, and, most of all, it was a
blast!” Teddy Ort – Ph. D. Student, Massachusetts Institute of Technology
(MIT)
“If University were like learning how to play a new instrument, where lessons
are the exercises and exams the final auditions, Duckietown would be the full-
blown rock concert, where you play for your fans and look to your heroes with
admiration.” Gioele Zardini – Ph. D. Student, ETH Zurich
“Spending the summer in Duckietown at MIT made me discover a completely new
world: I understood that education can be a game and learning can be fun!”
Valeria Cagnina – Young Enterpreneur
“The Duckietown class is the autonomous driving pie: the filling is hardcore
robotics, the casing is artificial intelligence, and as a plus, you get some
funny ducks on top!” Manfred Diaz – Ph. D. Student, University of Montreal
Testimonials
VOICE OF THE EXPERTS
“Starting from MIT CSAIL, Duckietown has grown into a global initiative that
is inspiring students around the world to learn about self-driving cars, as
well as the science and engineering of autonomy.” Prof. Daniela Rus –
Director, Computer Science and AI Lab (CSAIL), MIT
“Teaching the Duckietown class was a wonderful experience for me and my
students. The materials are great and the hands-on experience with the robot
really helps reinforce the curriculum.” Prof. Matthew Walter – Toyota
Technological Institute in Chicago (TTIC)
“It is great to see Duckietown host the AI Driving Olympics at ICRA and
N[eur]IPS. What a fun way to demonstrate the real challenges in building and
deploying self-driving cars!” Prof. John Leonard – Massachusetts Institute of
Technology
The AI Driving Olympics is a great way to push the limits of deep learning on
physically embodied systems.” Prof. Yoshua Benjio – 2018 ACM A.M.Turing award
winner, University of Monreal
Read User Manual Online (PDF format)
Read User Manual Online (PDF format) >>