Design of a Hyper-Casual Futsal Mobile Game Using a Machine-Learned AI Agent-Player

Mobile games continue to gain popularity, and their revenues are increasing accordingly. However, due the inherent constraints of small screen sizes restrictions computing, it has been considered challenging simulate complex gameplay soccer games. To this end, paper aims design develop a simplified version five vs. hyper-casual futsal game with only three player positions: goalkeeper, striker, defender. It also tests demo verify whether is possible implement an AI agent−player for each position machine-learn run on mobile device. A was simulated using both PPO SAC algorithms, feasibility stability algorithms were compared. The results showed that achieved assigned objectives successfully machine-learned. When compared, algorithm more stable state than when directed interactive techniques. This shows great potential application machine-learned agent−players simulators platforms.