Day 3: The Swedish domination. Follow Sagar's adventures as he hangs around the Netherlands in the days leading up to the Grand Cooperative Driving Challenge.
Heretic people work on Sundays too, even though the Hotel Breakfast is not available early in the morning because everyone else is taking it easy. So we trooped out of our rooms and picked up little cardboard boxen filled with random food (a la McDonalds Happy Meal) and ate it in the car on the way to TNO.
The day started off with qualification tests for the event. These are tests that each team needs to pass to prove that their vehicle is safe enough to compete. TNO had closed off the section of the highway and it was here that the tests took place. It was fun to see the measures TNO had in place to ensure safety. Not only were there a couple of men manning each exit but there was a big firetruck in case someone crashed and burned literally, rather than metaphorically. First off was the German team from the University of Karlsruhe and in typical German fashion, they smoothly and flawlessly moved from one test to another and qualified with ease. That naturally upped the stakes for the next team, which was us. We breezed through the qualification as well and you could see the TNO guys were quite pleased. At the end of the first day's qualifications, only 4 teams turned out to be ready enough to be authorized for the first automatic platooning test. These were the Germans and the three Swedish teams (KTH, Chalmers and Halmstad). The rest of the teams are in various levels of readiness and we hope that by the end of the week, all will pull together and we will have tough competition.
A remarkable observation was that almost all the teams have individually evolved similar architectural patterns. The core of the system is a Linux based computer connected to an environment capable of running Simulink control models. Connected to this are peripherals (RADAR, LIDAR, GPS..) that assist in decision making. This might point to the conclusion that for the specified constraints that we operate under, this is an optimal architecture pattern. Of course, this conclusion is logically fallacious and it might be the case that geeks think alike.
TNO wants to document this week's work efforts and there are people with cameras and videorecorders and mics moving around. It is a new experience to be working with camera flashes and pausing work to do the occasional interview. As a matter of fact, after my first interview, I realized that I was standing in Team FUTURUM's pit. You are welcome, guys :)
In the afternoon, we drove twice in manual platoons around the actual highway section where the final event will take place. After that, the 4 ready teams tried an automatic platooning drive. We started off fairly well, but had some issues with the speed limit being (not) broadcasted due to which we came out of automatic mode and could not return to it.
Since things are going well at the moment, we've turned our attention to some eyecandy and PR. Simon is developing a nice graphical view of the system complete which will give us a one screen overview of our local surroundings. This will show the detected cars, their orientation and speeds, the platoon states and so on. If it works, we will get the film team to record it during a live test and hopefully it can form a part of the final video. Doing something is good; doing it in style is even better :D In parallel work tracks, Henrik is working on the controller, Dennis and I will continue with minor system corrections for issues we've noticed from time to time but which were low on the priority list.
We look forward to an interesting day tomorrow!