Rational interaction in autonomous systems

Many emerging applications of AI require interaction, cooperation, and coordination between machines. Within this context, this research is aimed at building multi-agent systems comprised of autonomous decision-making components. Each individual component must be designed to accomplish its own goals despite possible conflicts with the goals of others. Focus will be on modelling rational interaction and devising effective interaction mechanisms for enabling autonomous conflict resolution. The project will involve formal models from game theory together with machine learning, but the ultimate focus will be on real-world applications such as building a team of humans and manned/unmanned vehicles (UAVs), or a team of robots in which each robot must adapt its behaviour in accordance with the behaviours of the others. The research involves theoretical modelling as well as simulations.