Home Work Resume

Master Thesis

Development of a combat scenario between two AI groups using tactical decision making

Game Engineering and Visual Computing
University of Applied Sciences Kempten

Artificial intelligence is an increasingly important and complex component of games. Characters make decisions and act on them. The cooperation of several AI agents as a group is relatively rare, especially outside of large game development studios. Therefore, the goal of this master thesis is the development of an AI for the control of groups using tactical and strategic information. It turns out that this is a relatively complex undertaking, but relatively good results and a clever artificial intelligence were achieved. The accompanying application was implemented in the Unity engine with the help of a number of other systems, such as influence maps, cover points, scripted actions or points of interest.

Cover System
This system allows ai agents to seek cover from enemy characters.
Scripted Actions
Certain areas, like buildings, get captured using predefined actions. This application uses scripted actions with squads of up to four soldiers to take houses in rapid fashion.
Influence Map
Influence Maps show military influence in certain areas of the map. Decisions of ai agents can be based on this information.
Group AI
The group ai controls every character of a team and uses all the available information to make tactical decisions and capture every point of interest of the level.
And many more features.

Video

Short demonstration of the finished application.