Hi guys, Sasha here! It’s been two months and we have been making nice progress. First of all, I am basically finished with calculating the ultimate raceline. There is just some small tweeking left, in order to make the program faster. This will be done by removing a lot of sample points on large straight sections of the track, for now every meter along the track is being analysed, which is really unnecessary on large straight parts (and there are relatively many of those). Having done that, with the details of the new track being announced, we are just some computing hours away from knowing our raceline, finishing what we started in my last blog.
That was it for the simulation, now the driving-style strategy itself. Like with the raceline, we shall use a genetic algorithm to find the best strategy. We shall start with different strategies, which we will rank using the simulation from earlier. Using the same process described in my last blog (with some adjustments discussed at the beginning of this one) we will hope to find the best strategy.
As for later, our plans are clear: Although this strategy and raceline which I discussed give a precise picture, it is mainly for ideal conditions, when no other cars are on the track. We however want to give the best possible strategy real-time, during the race. That means that we have to make quicker programs that can give a good result while including more variables. This will certainly be a challenge, but we have some nice ideas and I hope to get you guys up to date to them real soon!