Development of an intelligent agent capable of playing a fighting video game
DOI:
https://doi.org/10.32870/recibe.v13i3.379Keywords:
Artificla Intelligence, Videogames, Intelligent Agent, Proximal Policy Optimization, Reinforcement Learning, Self Play, ML Agents, UnityAbstract
Video games are an important tool as a testing ground for reinforcement learning algorithms; however, they are also a valuable tool for generating artificial intelligence for games themselves. This research was divided into two general phases: the development of a fighting game demo called Brain Fighter and the creation of a reinforcement learning model based on the algorithm known as Proximal Policy Optimization to train an intelligent agent capable of playing Brain Fighter satisfactorily.This paper presents the best model obtained after experimenting with six different models, and it also explains the methodology used to conduct the entire research, from the creation of Brain Fighter in Unity to the selection of the reinforcement learning algorithm available in ML-Agents, the creation of the model, the training, the results obtained, and the general conclusions of the work.References
- Biscontini, T. (2023). Deep reinforcement learning (deep RL). In Salem Press Ency-clopedia of Science. Salem Press.
- Community, Rlb. (2024). RLBot. https://rlbot.org/
- Gu, Y., Cheng, Y., Chen, C. L. P., & Wang, X. (2022). Proximal Policy Optimization With Policy Feedback. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 52(7), 4600–4610. https://doi.org/10.1109/TSMC.2021.3098451
- Juliani, A., Berges, V.-P., Teng, E., Cohen, A., Harper, J., Elion, C., Goy, C., Gao, Y., Henry, H., & Mattar, M. (2018). Unity: A general platform for intelligent agents. ArXiv Preprint ArXiv:1809.02627.
- Landers, M., & Doryab, A. (2023). Deep Reinforcement Learning Verification: A Survey. ACM Comput. Surv., 55(14s). https://doi.org/10.1145/3596444
- OpenAI. (2018, April 27). Gym Beta. Https://Openai.Com/Research/Openai-Gym-Beta.
- Sarrià Pascual, D. (2022). Development of a competitive Rocket League bot using reinforcement learning. Universitat Politècnica de Catalunya.
- SaturnCloud. (2023). Self-Play in Reinforcement Learning. Https://Saturncloud.Io/Glossary/Selfplay-in-Reinforcement-Learning/.
- Sutton, R. S., & Barto, A. G. (2018). Reinforcement learning: An introduction. MIT press.
- Tantawi PhD, R. (2023). Machine learning. In Salem Press Encyclopedia. Salem Press.
