Improving procedural 2D map Generation based on multi-layered cellular automata and Hilbert curves

Procedural Content Generation (PCG) for Games is a field that in the past few years has seen both extensive academic study and practical use in the games industry. One of its common uses being the generation of maps for levels within games that rely on replayability. While Cellular Automata is a PCG technique widely used for the creation of minor graphical systems, it has not yet seen much practical use in the generation of levels, part due to its inherent stochastic nature. With the purpose of presenting a malleable approach for improving levels created through Cellular Automata, this work presents a methodology that guides the generation process through the use of fractals, specifically Space-filling Curves. The product Automata of this process are implemented and polished on the Unity game engine, as to present their potential for generating procedural levels. Results show that this methodology can be used for the generation of organic, cohesive game levels.