Project Wingman

The technical merit of the Project WingMan solution lies in its integration of advanced AI technologies for real-time data analysis, performance assessment, and retraining capabilities. By leveraging machine learning algorithms and Large Language Models (LLMs) trained on FAA standards and thousands of hours of flight data, the system can evaluate complex flight data streams, including avionics, audio, and video, to deliver precise and actionable feedback. Project WingMan's unique capability to assess stress levels and human factors based on cockpit actions and voice patterns adds an innovative layer of insight into pilot performance. Furthermore, its ability to automatically identify areas of weakness and retrain pilots in Immersive Training Devices (ITDs) ensures targeted, data-driven learning that continuously improves proficiency. The solution also scales efficiently, reducing instructor workload while maintaining high standards of training, making it highly adaptable for large-scale pilot training programs such as those in the Air Force. By combining real-time flight data analysis with voice-based interaction, Navi's AI delivers personalized, dynamic training experiences that evolve based on each pilot's performance. This AI-powered instructor provides immediate, context-aware feedback by analyzing avionics data and cockpit audio, creating an interactive, conversational approach to flight training. With this unique integration of generative AI, Navi transforms the learning process, enabling pilots to not only correct their mistakes but also deepen their procedural understanding, ultimately enhancing both safety and proficiency in the cockpit.  This includes:(1) Interactive Ground training:  AI instructor interacts with and adapts content to student learning style, replaces all books and outdated material and the need for 100+ classroom hours. Improving comprehension and retention. Pilots can train anywhere, stay current and save time.(2) Significantly reduce Flight Time:  Navi teaches procedures, avionics interfaces and ATC communication outside the cockpit, allowing for more repetition in a simulated environment. Keeps students proficient while on grounded due to airplane and instructor availability. Reducing need for retraining.(3) Standardize/Improve Training:  Navi tracks progress by streaming and analyzing cockpit audio and avionics data, automating flight debriefs and adjusting syllabus to student progress. Structuring prep sessions to ensure students exceed in each flight lesson. Based in San Francisco, the Navi team consists of flight instructors, pilots, and AI engineers from Carnegie Mellon, MIT, IIT, Google, and Facebook with a wealth of experience from previous collaborations with Boeing, the FAA, the US Navy and DARPA.