Machine Learning and Artificial Intelligence in Air Traffic Control: Recent Development as a Research Perspectives

Abstract

Air Traffic Control (ATC) is performed in a centralized way by human controllers, which monitor and manage all aircraft flying through the airspace sector under their responsibility. In extremely dynamic and unpredictable circumstances, ATC is a safety critical supervisory mechanism that operates in real time. Due to the rapidly increasing complexity of air traffic in both low-altitude (drones and eVTOL aircraft) airspace and traditional (commercial airliners), an independent air traffic management arrangement is required to make available high density air traffic while ensuring airspace separation that is safe. The most difficult responsibility entrusted to controllers is to avoid crashes, design aircraft paths, and ensure that these flights take off and arrive at their destinations on schedule. The competences of Artificial Intelligence (AI) that might be implemented in these controls to make rapid judgments and eliminate the need for human interaction are discussed in this study. It has been declared effective in providing timely decision-support guidance in a variety of real-world situations that might be life-threatening. The safety of present air traffic control systems is a key stumbling block to automation. The currentstudy examines machine learning and AI utilization to automate an existing ATC system. ML is an area of AI concerned with the creation of algorithms which could be programmed to learn from prior data. This paper provides an overview of current breakthroughs in aviation using AI and Machine Learning (ML) algorithms to address concerns such as automation transparency, air traffic, airline delays and reduction.