Autonomous systems are quickly evolving to provide a versatile and essential capability in both military operations and commercial applications. From a human-systems perspective, these recent technological developments are changing the role of human operators into that of supervisory controllers of complex automated and autonomous systems who must maintain situation awareness (SA), and be ready to rapidly intercede in complex or critical situations that require human judgment and general intelligence. Unfortunately, this rapidly advancing technology has exceeded the ability of traditional methods, often relying on expertise and intuition, to predict how operators will perform and interact. In support of U.S. Navy unmanned aircraft system (UAS) airspace integration initiatives, a high-fidelity simulated representation of air vehicle operator (AVO) behavior and performance is in development. The Operator Model (OM) employs machine learning (ML) and other artificial intelligence techniques for characterizing observed responses of AVOs to air traffic encounters, along with a means to reproduce and generalize those responses for use in faster-than-real-time constructive computer simulations. In doing so, this model addresses several needs, such as providing an economical means of generating the volume and variety of human-performance data required for platform certification, and informing future design and training decisions. The OM represents a significant new capability for unmanned aviation-systems development. It combines task analysis and experimental psychology with advances in machine learning to support simulation-based acquisition in a complementary and cost-effective manner, enabling certification of defense systems with higher levels of autonomy and more complex patterns of human-computer interaction (HCI). This paper will provide an overview of the OM hardware and software architecture, and highlight the Live-Virtual-Constructive (LVC) trials that have been performed at the Naval Air Warfare Center, Aircraft Division (NAWCAD) to validate UAS sense-and-avoid (SAA) capabilities. Phase 1 results indicate a strong agreement between LVC and OM measures of effectiveness (MOEs).