Conversational Agents in a Pattern Oriented Training Environment (CAPOTE)

Training programs such as Specialized Undergraduate Pilot Training (SUPT) present students with an array of complex skills to integrate in a dynamic, time-sensitive performance context. Instructors are frequently forced to spend some of a student's limited flight hours reviewing skills that could be trained offline, such as situational awareness (SA), basic pattern geometry, and radio communications. Training could thus be made more efficient and effective if students had access to simulation-based training devices for learning and integrating pattern geometry, radio communications, and other SA-related skills. This program addresses the need for greater access to advanced training devices with an automated intelligent tutor that reduces reliance on human instructor pilots and allows on-demand practice using low-footprint, portable devices. We propose Conversational Agents in a Pattern Oriented Training Environment (CAPOTE), a framework that offers important, high-payoff extensions to COTS simulation. CAPOTE employs intelligent tutoring to provide effective, instructor-less training opportunities, and synthetic agents that can assume key roles within training scenarios and interact in spoken language with the trainee. The CAPOTE framework will guide the development of a technology prototype for SUPT operations in the T-6 that can be deployed to UPT training bases for testing and evaluation.