This is an A.I. test that can complete any track you create. It uses generations and optimizes its path based on successes from the last generations. (Full explanation below :))
How does this project work? The main idea of a genetic learning algorithm (GLA) is to use a generation-based approach to tackling a problem. The main principle involves succumbing many different "clones" of a certain type to try to accomplish the problem themselves. This often includes random elements so that there is a chance that a "clone" in a generation makes a breakthrough. In this project, each generation contains 50 "cloned cars" that each attempt to follow the previous generation's footsteps (with a random element, explanation above) and then try to make its own breakthrough. When a specific "clone" makes it farther than the group, its data is recorded and used by the next generation. When all clones "die," they go back a bit and attempt to accomplish the problem again. If you look inside, you'll see that the clones have no color detectors to "follow a path," they just expend many hundreds of cars (in the name of science) every time you run the project :) #ai #a.i. #genetic #learning #algorithm #car btw. if you made it this far comment "I know what a GLA is :)"
