Context-Driven Autonomy for Enhanced System Resilience in Emergent Operating Environments

As higher degrees of operational autonomy are sought in engineered systems, resilient command and control (C2) capabilities are essential for mitigating performance degradation and disruptions. Autonomous C2 systems typically excel in performance optimization and failure-prevention within known or anticipated environments, but can exhibit poor efficacy given a lack of preemptive knowledge of needs and environments. This paper presents a context-driven decision engine for resilient C2 of engineered systems in emergent operating environments. The solution integrates diagnostic and prognostic heuristics to 1) assess system state of health (SoH) based on operational availability; 2) establish situational awareness; and 3) drive C2 decisions based on scenario-specific constraints and priorities. The solution is applied to an air traffic management problem for autonomous aircraft operations based on a Federal Aviation Administration Next Generation Air Transportation System (NextGen) concept. Agent-based modeling and Monte Carlo simulation are used to evaluate the incident rate (lambda) and mission reliability (R-M) of the proposed context-based solution against rule-based and utility-based C2 solutions, including the Radio Technical Commission for Aeronautics DO-185B standard and Dynamic A-star Lite (D* Lite) algorithm. Paired t-tests are used to compare approaches based on lambda and R-M given identical test cases. Results suggest that the proposed context-based solution can concurrently enhance threat avoidance and adaptation capabilities in emergent environments, whereas unilaterally optimized performance is observed via the rule-based and utility-based C2 solutions.