CCC--Uno AI agent auxiliary system
Chi Zhang, Zhengxiao Chen · Applied and Computational Engineering · 2024
In this paper, we explore the design and implementation of a UNO game AI Agent based on the Mini-Max algorithm, and we named it "CardCraft Champ". UNO, as a popular card game, is filled with strategies and uncertainty, making it challenging to design an AI Agent that can efficiently handle different game scenarios. Our goal is to develop an AI system that can intelligently select cards, respond to opponents’ strategies, and achieve victory in UNO games. The core idea of "CardCraft Champ" is to use the Mini-Max algorithm and Alpha-Beta pruning for decision-making. By conducting deep searches and evaluations on the game state, our AI Agent can predict various possible game progressions and make optimal decisions based on expected outcomes. The Alpha-Beta pruning technique can accelerate the search process even further, enabling our AI to find the optimal solution within a limited time and make wise decisions in the game. Our research not only focuses on the implementation of the algorithm but also on its application to actual UNO games. Through comparative experiments with other existing UNO game AIs, we demonstrate the superior performance of "CardCraft Champ" in different scenarios. The experimental results show that our AI Agent can minimize risks in the game while maintaining a high winning probability against various opponent strategies and game scenarios. Additionally, we consider the scalability and applicability of "CardCraft Champ". Our system performs well in battles of different difficulty levels, proving its adaptability and stability. We also explore possible directions for improvement, such as introducing deep learning techniques to enhance the AI's decision-making capabilities.