This project is organized in the subject of CT7121 Machine Learning and Data Analytics. The purpose of this project is to study the use of Machine Learning techniques and AI development through Unity ML-Agents, focusing on the comparison between traditional AI systems such as NavMesh and Machine Learning systems.
In this project, there is an AI character or Player AI that is Machine Learning trying to survive in the game environment by collecting items and eliminating enemies. The AI receives data through sensors or the 3D Ray Perception Sensor to help decide on movement and attack, with a reward and punishment system.
This project uses the PPO (Proximal Policy Optimization) technique and the data is checked through TensorBoard. This entire project is developed in Unity in a scene named "projects" and all projects are in the branch named "ml-agents".
(This is a scene illustration showing the AI machine learning in action, picking up items and facing the enemy to shoot.)
