Links

Here you can find a collection of helpful links related to the development of Self-Driving Cars. This site is still heavily under construction ;).

Self Driving Cars

Self-Driving Car Engineer Nanodegree @ Udacity
The likely best way to start your career as Self-Driving car engineer and to build a foundation of knowledge required for developing autonomous vehicles in general.The course is separated into three terms with a length of three months each, so you can finish it within about 9 months of time. You are allowed to pause between the terms (and many do, for example for holidays), but I also already met people who finished it in 8 months instead of 9.

https://www.udacity.com/course/self-driving-car-engineer-nanodegree–nd013
Intro to Self-Driving Cars Nanodegree @ Udacity
A really awesome 4-month course to get a first insight into the awesome world of Self-Driving cars and all the difficulties a Self-Driving car engineer has to master.

https://www.udacity.com/course/intro-to-self-driving-cars–nd113

Courses

The Unreal Engine Developer Course – Learn C++ from scratch
I myself just started this course a couple of weeks ago. It has more than 55 hours of video material and more than 314 lessons. With a rating of more than 4.7 stars this is right now simply the course on the web, if you want to get lots of practical experience using C++ and handling logic in a continuous, 3 dimensional world. And next to all of this the result of all your exercises after finishing this course is an awesome portfolio.

Important note: In case of Udemy you can for the most courses assume at least a factor 4 of the video time as time requirement to finish this course, so in this case 55 hours * 4 = 220 hours = easily two months of dedicated full time learning. As already mentioned above this is a great course if you have the time and want much experience for less money, not if you want to learn the basics of C++ quickly :).

https://www.udemy.com/unrealcourse/

Krista King Math
I “paused” *cough* high-school (meanwhile nearly twenty years ago), so I missed the final parts of Calculus (Analysis in German) which are required for basically every Deep Learning based application. As I also wanted to understand the details I needed to continue this journey and refresh the parts already learned. One very often recommended way is Khan Academy where you can learn everything for free and are guided by quizzes. If how ever 30 $ a month don’t hurt you and you want awesome courses and to be guided professionally and with the possibility to ask a real human for details you should choose Krista King:

http://www.KristaKingMath.com

or at Udemy (no personal support, just the great videos and cheat sheets):

Multivariable Calculus

https://www.khanacademy.org/math/multivariable-calculus

Linear Algebra

https://www.khanacademy.org/math/linear-algebra

Standford Online’s Machine Learning Course @ Coursera
A very theoretical/math intensive course based on Octave/Matlab. Very recommendable if you also want to understand the details of the “black box”.

https://www.coursera.org/learn/machine-learning
Machine Learning Engineer @ Udacity

https://www.udacity.com/course/machine-learning-engineer-nanodegree–nd009t
DeepLearning.ai @ Coursera
Definitely one of the courses at the absolute top of my list and (at least I have heard so) a very good extension to the DLF Nanodegree by also presenting techniques such as YOLO.

https://www.coursera.org/specializations/deep-learning
Artificial Intelligence Engineer @ Udacity

https://www.udacity.com/ai
Deep Learning Foundation Nanodegree @ Udacity
I’m right now in the final phase of this course and it was awesome, because you got in touch with nearly every important type of Deep Learning network and taught it’s details. Unfortunately it is not as SDC focused as I had hoped, so for example features like semantic segmentation or in general localization of a detected object class were missing and need to be googled in the web. How ever, it’s called “foundation” for good reason, because it does give you a great foundation in Deep Learning :).

https://www.udacity.com/course/deep-learning-nanodegree-foundation–nd101

Articles

A great collection of articles about Deep Reinforcement Learning by Moustafa Alzantot

https://medium.com/@m.alzantot

Keon’s awesome straight-forward DQN solution
https://keon.io/deep-q-learning/
Richard S. Sutton’s and Andrew G. Barto’s Introduction to Reinforcement Learning

http://incompleteideas.net/book/bookdraft2017nov5.pdf

Data Science Specialization @ Coursera
https://www.coursera.org/specializations/jhu-data-science

Many of the projects and exercises of Udacity’s Self-Driving Car Nanodegree
https://github.com/udacity/self-driving-car
Udacity’s Deep Learning exercises and projects
https://github.com/udacity/deep-learning
Uwe Sterr’s GitHub repo with some nice exercises about SDC and Deep Learning
https://github.com/uwesterr
Alyxion’s (my) GitHub repo with several demos and exercises about Deep Learning and SDCs
https://github.com/alyxion