Autonomous cars are designed to make driving a better experience by minimizing human errors that might be fatal. There are various levels of Automation implemented in cars over the past few decades. In a completely automated car, making the machine recognize and learn its surroundings is just as challenging as it is to differentiate between the lane’s edge and an obstacle on the lane. This is primarily achieved with the help of various sensors that detect and collect various types of data for the machine to learn from. This model is built in 2D by simulating an environment that allows the car to learn through a feed-forward neural network. The car can detect and stay within lanes and can also navigate through traffic. After several epochs with multiple generations of cars, the machine learns the art of driving.