Combining coordination with usage policies to improve mission scheduling resilience

摘要

Intelligence, surveillance, and reconnaissance (ISR) assets must coordinate to maximize their availability, timeliness, and accuracy of information in a fast-paced, everchanging, and often adversarial environment. The distributed constraint optimization problem (DCOP) is a canonical formulation for representing coordination problems found in the planning and collection phases of global integrated ISR operations. However, ISR assets are often owned by different organizations and have ISR usage policies that dictate additional constraints based on organizational needs. The distributed, agent based system presented in this paper seeks to add resilience to ISR missions by combining DCOP techniques to optimize the scheduling of ISR assets while using policy negotiation to inform the assets??? usage policies in order to increase the number of tasks performed by the system. We overview the system architecture, the coordination protocol, the usage policy engine, and the negotiation strategies used. Broad experimentation shows that use of schedule coordination coupled with policy negotiation outperforms the use of either technique alone.