Pole Balancing

Neural networks evolve to balance a pole on a moving cart by applying force in either direction.

Generation0

Best Fitness0

Alive Count0

Best Genome

The best genome appears here after the first generation.

About This Simulation

This simulation demonstrates the NEAT algorithm applied to the classic pole balancing problem. Neural networks evolve to balance a pole on a moving cart by applying force in either direction.

Technical Details

Neural Network Inputs: Cart position and pole angle
Neural Network Outputs: Force direction (push left or right)
Fitness Function: Based on time the pole remains balanced
Failure Conditions: Cart moves too far (±2.4m) or pole angle exceeds ±12°