An Analysis of Finite State Machine Based Enemy Artificial Intelligence in Kirby: Nightmare in Dream Land

The type of AI used to design video game enemies greatly affects the gameplay experience of speed, difficulty, and enjoyment. In most cases, the majority of developers who create 2D platforming games will choose to implement a simple but efficient AI design over an advanced AI model that learns based on experience. One of these simpler AI models that is frequently utilized by 2D platforming game developers is the Finite State Machine (FSM) model. The FSM model creates an organization of the enemy's actions into a limited number of well-defined behaviours, while also indicating how these behaviours relate to one another. We look at how AI uses the FSM method in the 2D platform game "Kirby: Nightmare in Dreamland," which first came out on the Game Boy Advance. The analysis of FSM models enemy AI behaviors and how those behaviors change and when they do so affects how hard the game is. Simulated experiments were conducted on how state time is spread out and how to make things harder by changing the attack cooldown. The results show that FSM-based AI is easy to control, doesn't need a lot of processing power, and has behavior that can be predicted. This makes it a good choice for platform games that are easy to get into. The results show that FSMs are still important in AI research and game design today.