Solution Concepts and Performance Guarantees in Partial Information Multi-Player Games

(Q rep reqd) This is AOARD US funding for $390k US dollars. In this project, we will build a novel framework for modelling and solving multi-team/player games with asymmetric information structures where the actions, states, and the pay-offs of different players and even the environment that they operate in are not universally known. We particularly focus on dynamic/differential games in complex networks with asymmetric information, where players uncertainties about the network and environment are integral parts of the approach. We aim to deliver robust and perception-aware team autonomy architectures and algorithms that have guaranteed performance, degrade gracefully under nonnominal conditions, and are implementable from a complexity perspective.