Skip to content
This repository has been archived by the owner on May 12, 2021. It is now read-only.

Tech-Stack: Java. An EA evolved the weights of a neural network to land spaceships.

Notifications You must be signed in to change notification settings

AverageHomosapien/Lunar-Lander-Evolutionary-Algorithm

Repository files navigation

CI-Coursework

This was a 3rd year coursework done as part of the Computational Intelligence module.

Overview

The task was to try to safely land spaceships in a game environment, using an Evolutionary Algorithm. A neural network was used by the spaceships to understand when, and when not to fire their thrusters. We were given the task of creating an evolutionary algorithm which evolved the weights of the neural network to allow spaceships to land safely.

I used a unique approach, using mutation as an exploration and exploitation operator and was able to achieve solid results for the task.

I give an example screenshot of the game environment below:

lunar-lander-game-environment

About

Tech-Stack: Java. An EA evolved the weights of a neural network to land spaceships.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published