CartPole Fuzzy Control System

A Mamdani-type fuzzy logic control system that balances the classic CartPole problem using continuous force control instead of the usual discrete left/right actions.
The controller reasons over four inputs — pole angle, angular velocity, cart position, and cart velocity — and applies a continuous force through 24 fuzzy rules, combining pole-angle stabilization with drift correction via an integral term.
It is built in Python on a custom continuous CartPole environment (Gymnasium classic-control) using scikit-fuzzy, with visualization of the membership functions and 3D control surfaces.