[This article belongs to Volume - 56, Issue - 01, 2024]
Gongcheng Kexue Yu Jishu/Advanced Engineering Science
Journal ID : AES-30-01-2024-02

Uma Perumal

Abstract :

Incorporating Machine Learning (ML) methods into flight instruction is now seen as a game-changing strategy to update aviation curriculum and raise safety benchmarks. In this work, many ramifications of using ML in pilot training, including the opportunities, threats, and ethical issues that may arise has been investigated. The research technique is methodical, consisting of steps like gathering data, cleaning it up, choosing an algorithm, building a model, and testing it. The results were shown using fictitious datasets whose parameters were based on those utilized in aviation. Predicting critical flying circumstances using the Isolation Forest method and evaluating pilot stress using neural network-based stress level prediction were the two key goals pursued. The results demonstrate ML's usefulness in spotting outliers and tailoring approaches to stress management. The ways in which these discoveries might improve aircraft safety, pilot happiness, and realism in training are emphasized in the discussion. The results of the studies are strengthened by the use of theoretical models like the Yerkes-Dodson law. In the publication, the author stress the importance of the research for future pilots, schools, authorities, and business executives. With its potential to reinvent aviation education, redefine operating standards, and eventually contribute to safer skies, machine learning has a place in the training of pilots. This study paves the way for more investigation into such emerging topics as real-time adaptability, extensive biometric integration, simulated reality experiences, and moral concerns.