Each year for the last four years I’ve worked on a major project relating to self-driving cars. In 2013 and 2015, I did the Simulated Car Racing competition (simulated driver would be more accurate). In 2014, I made my drive by wire model car and started learning about OpenCV. In 2016 I made a special effort to learn and apply OpenCV. When I saw that Udacity was offering a premium course for Self-Driving Car Engineering, I knew I had my project for 2017.
Although some people think the market for Self-Driving Car Engineers is strong (and some people think it is ridiculously strong!), I’m not necessarily counting on that being true. For me the goal is to gain as much useful knowledge as possible to continue to do whatever I can to hasten the demise of cars being controlled by humans who are staring at their phones.
I’ve now completed about half of the first 3 month term. Of course I have a lot to say about this course (mostly good), but I’m going to hold off for now. So far it has been challenging but pleasantly so (I finished 6 weeks of lessons in 4 weeks). One thing I’m very pleased about is getting past the starting blocks with OpenCV. I did learn quite a bit about OpenCV last year, but I was constrained by implementation considerations. With this course, the industry standard set up is established and then provided for. This may seem trivial, but for me it was the removal of a serious barrier to entry. Being able to just get on with actually working with OpenCV is proving to be quite empowering.
I don’t like to say too much about my accomplishments before I accomplish them so if I manage to get through this program I’ll have much more to say about it. Until then I’ll try to occasionally post some of the interesting things I’ve been working on. Although I have many ideas, I don’t know exactly how I’ll apply what I’ve learned so far in the course, but I can say I’ve already got my money’s worth.