Adaptive NPC behavior in video games with AI for real-time player interaction
Salek, MD Peash Been Salek (2025)
Kandidaatintyö
2025
School of Engineering Science, Tietotekniikka
Kaikki oikeudet pidätetään.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi-fe2025080881520
https://urn.fi/URN:NBN:fi-fe2025080881520
Tiivistelmä
This bachelor's thesis explores the NPC behavior in games, specifically, how NPCs function as an AI driven agent. The primary objective of this study is to validate data efficacy and examine use case studies of NPC implementation in AI. From the early days to the modern era, NPC action in games changed a lot thanks to the help of modern game engines like Unity and Unreal engine, which have AI tools and features. These tools and feature help developers to create an interactive and efficient NPC.
In this development and implementation process, it was observed that machine learning techniques also evolved. By using more powerful machine learning tools, developers can make AI driven NPC. Using behavior tree, reinforcement learning, Q-learning algorithm, path finding, AI NPC become more flexible, interactive, dynamic and adaptive. Some games already include AI driven NPC, which guide the player in better path, help them to complete the quest, communicate like human – both in thought and emotion.
The result demonstrates that, while modern tools can greatly simplify the game development process there are also challenges, such as when creating an artificial intelligence non player character. Developers must also be careful in game balancing because of the powerful process. This thesis offers the importance of AI NPC in the game, the role of game engine and machine learning in creating AI Driven NPCs and their behavior, and how to creatively leverage modern tools in NPC development.
In this development and implementation process, it was observed that machine learning techniques also evolved. By using more powerful machine learning tools, developers can make AI driven NPC. Using behavior tree, reinforcement learning, Q-learning algorithm, path finding, AI NPC become more flexible, interactive, dynamic and adaptive. Some games already include AI driven NPC, which guide the player in better path, help them to complete the quest, communicate like human – both in thought and emotion.
The result demonstrates that, while modern tools can greatly simplify the game development process there are also challenges, such as when creating an artificial intelligence non player character. Developers must also be careful in game balancing because of the powerful process. This thesis offers the importance of AI NPC in the game, the role of game engine and machine learning in creating AI Driven NPCs and their behavior, and how to creatively leverage modern tools in NPC development.