Developing a Unity component that implements planning and plan recognition for use in games.
Games have long been an important application area for AI research. The majority of this research has been focused on improving the play skill of AI agents against human players. However, there are increasingly many video games in which several players have to cooperate, and AI agents are notorious for lacking collaboration with human players. In this project, we will build on previous research into cooperative agents to develop a generic plugin component for the game engine/middleware Unity that can be used in a variety of games. Our project will make use of well-known techniques such as planning, which allows the agents to find action sequences that need to be executed to reach a goal, and plan-recognition, which allows the agents to detect what their collaborator is attempting to achieve. We will implement these techniques for use in Unity, and perform experiments to determine their efficiency and effectiveness.
Pablo Sauma Chacón
Unidades académicas colaboradoras