Self-Driving Car Simulation Model

The model has five input neurons (sensors), four output neurons (actions), and some hidden layers too. I have used the technique called NEAT (NeuroEvolution of Augmenting Topologies) to train the model using the Fitness Metrics (a fitness of a car increases depending on the distance it covers without crashing). The result was quite interesting, after 11th generation most of the cars were able to survive