Cooperative Decision and Control with Intermittent Asynchronous Communication

Communication rates in UAV operations can be poor, intermittent, and asynchronous due to mitigating factors within the operating environment such as equipment limitations, equipment configurations, terrain effects, and weather conditions. The objective of this research effort is to demonstrate the coordinated control of multiple UAVs in a closed-loop, dynamic framework using limited information and intermittent asynchronous communications. This objective will be accomplished through a Decision Architecture for UAVs (DA-UAV) that is distributed in execution and functions without reliance on a centralized controller. In this proposed research effort, the DAC/GMU Team will implement a distributed inference architecture using a framework called Multi-Sectioned Bayesian Networks (MSBN). Using this approach, it is possible to decompose a unified Bayesian Network into subcomponents that operate at different locations, while intermittently exchanging data to maintain global consistency. Furthermore, the DAC-GMU team will implement different coordinated control algorithms and a Value of Information (VOI) technique to enhance selected coordinated control methods. Using VOI, actions are ranked based on their ability to effect the overall mission value of the overall system. Within this methodology, the actions of individual UAVs are geared towards avoiding redundancy within the system, causing the overall system to behave more efficiently.