Explainable Artificial Intelligence to improve air traffic management
Predicting air traffic, optimising traffic flows and decision-making are stressful and difficult tasks for operators in the ATM (Air Traffic Management) area. The aim of the Mälardalen University (MDH) research project ARTIMATION is to assist the ATM operators.
The research project ARTIMATION focuses on investigating AI (Artificial Intelligence) methods, based on the domain of XAI (Explainable Artificial Intelligence), in predicting air transportation traffic and optimising traffic flows. Safety is the major pillar to air traffic management, and no black box process can be inserted in a decision-making process when human life is involved.
XAI refers to methods and techniques in AI where the results of a solution can be understood by humans, and hence be useful to human operators with different work-related tasks. It contrasts with the concept of the “black box” in machine learning where even its designers cannot explain why an AI arrived at a specific decision.
– In this case, information is being analyzed and then presented by transparent AI models and explanation decisions provided by AI systems, says Shahina Begum, Professor, Deputy Group Leader of the research group Artificial Intelligence and Intelligent Systems at MDH.
It is a process which will be useful to end-users such as ATM operators, working with air traffic management – predicting air traffic, optimising traffic flows and decision-making. These are complex and stressful tasks involving a lot of processing of content and information, which are being done manually today. The transparent AI models with explainable decisions provided by ARTIMATION will be helpful to the ATM operators.
MDH is coordinating the project and also leading a work package, called “Lifelong machine learning with human-centered AI”.
– Here, the aim is to investigate the applicability of AI methods from the domain of XAI that includes visualisation, explanation and generalisation with adaptability over time to ensure safe and reliable decision support. Again, we will investigate specific features such as lifelong machine learning, Data Driven Storytelling, Immersive Analytics, Visualisation, Brain-Computer Interface (BCI) and user centric AI to make the AI system very useful for operators of ATM-systems, says Mobyen Uddin Ahmed, Associate Professor at MDH.
The project uses human/end-user input while developing the AI system. This will improve the functionality, acceptability and trustworthiness of AI systems in general, but also leading to fulfillment of global goals such as improvement of industry, innovation and infrastructure in society.
Results from the project will be useful to AI researchers as well, who will benefit from the research in terms of transparency, generalisation and explainability of AI methods. Social systems and technology providers will also benefit from the results, hopefully leading to AI systems which will be more communicative and reliable by the human users.
The project has received funding from the SESAR Joint Undertaking (JU). The time schedule for the project is Jan 2021 – Dec 2022. Other project participants besides Mälardalen University are Deep Blue (an SME in Italy), Ecole Nationale de l’Aviation Civile (ENAC, aeronautical research institute in France) and the University of Rome “Sapienza” (UNISAP, Italy).
The ARTIMATION consortium members have been carefully selected to cover all the required research areas in a well-balanced way, utilising their expertise, prior collaborations, and state-of-the-art technical background to match the project’s objectives successfully.
Contact for scientific information:
For more information about the project, please contact:
Mobyen Uddin Ahmed, Associate Professor, Mälardalen University, Phone: +46-21-107369 firstname.lastname@example.org