A project at Mälardalen University (MDU) has led to the development of an AI system capable of calculating delays, predicting hazards, and identifying disruptions in the aviation industry. This system not only aids air traffic controllers in anticipating and conveying potential disruptions to passengers and aviation operators, but it also helps prevent these disruptions from occurring.
MDU researchers have created a prototype AI system designed for air traffic control (ATC) professionals working in Air Traffic Management (ATM). In ATM, it is crucial for operators to optimize traffic flows, identify possible aircraft collisions, and offer guidance to avert them. This AI system is a predictive tool that delivers optimal solutions to operators while also clarifying why they are optimal.
The system is developed within an Explainable Artificial Intelligence (XAI) framework, offering reliable solutions that are easily comprehensible for operators. It employs a 2D/3D map to display all feasible solutions, emphasizes the best choice, and supplies guidance to pilots or air traffic controllers to maintain safe distances between aircraft.
The AI system was developed as part of the Artimation project at MDU, with input from human end users. Mobyen Uddin Ahmed, Professor of AI at MDU, states that the project's results will enhance the functionality, acceptance, and dependability of AI systems in general while also addressing global objectives such as the enhancement of industry, innovation, and infrastructure in society.
Shahina Begum, Professor of AI at MDU, adds that the project's outcomes will be beneficial for other AI researchers, who can leverage the research in terms of AI transparency and explainability. Moreover, technology providers will gain from these results, potentially leading to AI systems that are more communicative and reliable for human users.
Src: Mälardalen University (Malardalen University)