Flappy Bird

Neural networks evolve to play Flappy Bird, deciding when to flap based on the position of upcoming pipe gaps.

Generation0

Birds Alive0

High Score0

Best Genome

The best genome appears here after the first generation.

Score0

Bird VelocityN/A

Bird Position YN/A

First Gap

Horizontal DistanceN/A

Vertical DifferenceN/A

Gap CenterN/A

Second Gap

Horizontal DistanceN/A

Vertical DifferenceN/A

Gap CenterN/A

About This Simulation

This simulation demonstrates the NEAT algorithm applied to evolving neural networks that can play Flappy Bird. Each bird uses a neural network to decide when to flap based on the position of upcoming pipe gaps.

Technical Details

Neural Network Inputs: Bird velocity, bird Y position, distance to gaps, gap vertical positions
Neural Network Outputs: Flap decision (yes/no)
Fitness Function: Based on survival time
Population Size: 300 birds per generation