UnityNEAT

UnityNEAT is a port of SharpNEAT from pure C# 4.0 to Unity 4.x (using Mono 2.6), and is integrated to work with Unity scenes for evaluation. UnityNEAT is created by Daniel Jallov as part of his master's thesis at the Center for Computer Games Research at the IT University in Copenhagen.

All the NEAT code is pure SharpNEAT, but is running in a single thread through Coroutines instead of using Parallel.For as in regular SharpNEAT.

Usage

To use NEAT in your Unity project, simply download the UnityNEAT folder and import it to your Unity project. In the folder you will find a few example experiments that you can copy and use for inspiration.

You will need to follow these steps to set up your scene to use UnityNEAT:

1. In your scene, add an empty game object ("Evaluator") and attach the Optimizer class to it. In the Optimizer script drag your Unit prefab to the Unit slot. Your unit is the thing which you are evaluating; a car, a robot, Pac Man, a colourful 3D shape. In this script you can set the trial duration (how many seconds each unit is evaluated) and the number of trials, which is useful if you use randomness.

2. Attach a script to your Unit prefab which extends UnitController instead of MonoBehaviour. Remember to use FixedUpdate instead of Update if the Time is scaled.

public class CarController : UnitController {
    bool IsRunning;
    IBlackBox box;

    void FixedUpdate() {
        ..
    }

    public override void Stop()
    {
        this.IsRunning = false;
    }

    public override void Activate(IBlackBox box)
    {
        this.box = box;
        this.IsRunning = true;
    }
    
    public override float GetFitness()
    {
        // Implement a meaningful fitness function here, for each unit.
        return 0;
    }
}

3. In the folder Resources there is a file called experiment.config.xml in which you can specify the parameters for the evolution, such as ComplexityRegulationStrategy and PopulationSize. See the original SharpNEAT project for more settings.

4. In the Optimizer script you can set the number of inputs and outputs in the consts NUM_INPUTS and NUM_OUTPUTS. The fields champFileSavePath and popFileSavePath are the locations where the xml file with the population and best individual is saved. By default it is saved to your local application path, which is in your User folder. The location will be printed in the console when run.

Example: Car experiment

In the Test scene folder you will find an example experiment with a simple car racing around a race track. In the folder CarExperiment you can find the CarController which calculates the fitness of the individual by measuring which road pieces the car goes through and how many laps it does. The car is penalized for hitting the walls.

A video of the evolution can be seen on youtube.

License

This project is released under the same license as SharpNEAT: http://sharpneat.sourceforge.net/licensing.htm which is the GNU General Public License, minimum version 3.

You should have received a copy of the GNU General Public License along with SharpNEAT. If not, see:

• www.gnu.org/licenses/
• GNU General Public License (GPL)