Tetris Solver: GA & Simulated Annealing

Sep 2025

Finished

This project implements a modern version of Tetris in Unity alongside two metaheuristic algorithms to autonomously solve piece placement: a Genetic Algorithm (GA) and Simulated Annealing (SA). The objective is finding the best arrangement for a sequence of pieces that minimizes the occupied board space. Each candidate solution encodes a sequence of piece movements. A custom fitness function evaluates the final board state. Both algorithms run extensive iteration-based experiments with configurable hyperparameters. A solution visualizer replays the best found placements in real-time during computation. The companion Python analysis module parses execution logs to generate detailed performance plots comparing GA vs SA across different configurations. A full research paper documents all implementation details, experiments, and decisions made.

AI

C Sharp

Jupyter

LaTeX

Matplot

Pandas

Plotly

Python

Unity

AI

C Sharp

Jupyter

LaTeX

Matplot

Pandas

Plotly

Python

Unity