Swarm Intelligence methods for multi-agent cooperation

OData support
Dr. Harmati István
Department of Control Engineering and Information Technology

My thesis presents the base concept of the most frequently used SWARM algorithms, the different ideas depend on the phaenomena of nature, and the explanation of the ability to optimize. After sketching the importance of these methods, we will focus on the Ant Colony Optimization, one of the most effective procedure capable of finding the shortest way in two dimensional graphs. This will serve as basis for the high-level military simulation. The next part of my thesis contains the description (program skeleton, graphics user interface) of the simulator (Art of War 3.0 © Peter Kisfaludi, Lajos Szarka) implemented in the MatLab environment, which is used in testing this method. The program is suitable for calling different strategies (exactly two in a game) based on different principles. The last section of the essay demonstrates a framework (developed by me) capable of constructing and synchronising high-level military strategies and tactical manoeuvres, which is suitable for testing in the simulator mentioned above. As a matter of fact, I realised a strategic intelligence with the method of SWARM optimizitaion in a layer based framework.


Please sign in to download the files of this thesis.