This week we are going to add a friend for the player. We’ll refractor the player creation code to reuse as much of it as possible and we’ll also create a new collision layer for the character and setup collisions between the player and his new friend.
In this series, we will write an AI that can teach itself to follow a set of rules. In an ideal world, I could write one AI character that could adapt to any environment within the game and provide an engaging experience for the player. While we are a long way from that ideal, this experiment may be a small stepping stone towards it.
In this part, we write the brains of our agents: a simple feed forward artificial neural network.
In this part, we will be implementing the neural network that we wrote in the last tutorial as the first step towards providing our little UFOs with some form of intelligence.
In the part, we start work on implementing a GA in our game to enable our UFOs to improve over time.
So far we’ve implemented our UFOs, provided them with neural networks, and started our genetic algorithm. Which means we’re almost finished the series. This week we’ll look at these topics:
We need a way of defining if a UFO is ‘fitter’ than another one.
We need a method of selecting the UFOs from the pool based on their fitness.
Once two UFOs are selected from the pool we need a method of combining their neural networks to create a new network.
The last in the series! We finish the genetic algorithm by applying a mutation to the newly evolved networks and then we write a bit of code to output the average time alive for each generation.