Risk-averse based paradigms for uncertainty forecast and management in differential games of persistent disruptions and denials

The present work includes the development of a multi-agent game-theoretic model and closed-form risk-averse strategies of sense and avoid for responsive sensor resources management. The model is intended to provide appropriate scenarios for teaming and cooperation of multiple sensor agents, targeting applications in surveillance, exploration and cooperative manipulation. The essential contribution is twofold. First, the robustness of distributed data collection is addressed by using innovative paradigms for performance uncertainty forecast and management. Second, the efficiency of sensing resources is enhanced by the realization of benefit and risk perceptions for tradeoffs between performance benefits and risks in conjunction of the attainment of risk-averse and efficient Pareto sense and avoid tactics despite of persistent disruptions from the uncooperative pursuers and active denials from the evasive objects. Finally, potential users of the model proposed herein are operations researchers with definite interest in exploring the capabilities, limitations, and performance envelopes of various decision strategies and algorithms.