Propelling the application of artificial intelligence in air traffic control

23/07/2021

Government backed grant announced for new five-year strategic partnership to help propel application of artificial intelligence in air traffic control.  

  • ÂŁ3m EPSRC funding will bring leading academic and air traffic management experts together 
  • AI integration aims to deliver safer, more efficient, and predictable air traffic control 
  • Targets data science systems that could work alongside and assist air traffic controllers 

The Engineering and Physical Sciences Research Council (EPSRC) has announced ÂŁ3m of Prosperity Partnership funding to support a five-year research programme between The Alan Turing Institute, the UK’s national institute for data science and artificial intelligence (AI) and NATS, the UK’s principal air navigation services provider.   

The vision of the first research project, named “Project Bluebird: An AI system for air traffic control,” is to deliver the world’s first AI system to control airspace in live trials, working with air traffic controllers to help manage the complexities of their role. This system utilises digital twinning and machine learning technologies and includes tools and methods that promote safe and trustworthy use of AI.  

The project is among several business-led Prosperity Partnerships announced today in support of the government’s ambitious new Innovation Strategy. The Prosperity Partnerships are supported with an investment of almost ÂŁ60 million by the EPSRC, part of UK Research and Innovation, businesses and universities.   

The Alan Turing Institute and NATS have been working together since 2018, applying data science to the complex airspace in the UK, highlighting a fundamental and specific need to build knowledge of how AI can add value to air traffic management. The collaboration has enabled a single team environment to embrace new research opportunities, such as data study groups and hackathons, a social event that brings people together, to explore novel thinking. 

The new strategic partnership has already commenced academic recruitment, pulling in a diverse mix of world leading interdisciplinary scientists, two Turing AI fellows, computer scientists, mathematicians, statisticians, engineers, experts in safe and trustworthy AI and computational ethics. This group comes together with the aim of building AI systems that recommend collaborative actions to operational air traffic controller teams. The objectives of the newly founded strategic partnership will explore the ethical considerations, feasibility, and trustworthiness of AI algorithms in safety-critical applications to produce a clear vision and deeper understanding of how AI could, one day, work alongside and support air traffic controllers.  

“Here at the Turing we are excited to build on our excellent foundational collaboration with NATS, moving from proof of concept and scaling our work to more ambitious targets. We know the challenges are great, for instance aviation’s commitment to net zero emissions by 2050, and we are working to tackle these together. Both NATS and the Turing’s values align around our commitment to open science and collaboration.  

We are working together to bring research around probabilistic digital twins to life, improving how we learn from data and design sustainable airspace, while also looking at ethical challenges of using AI in air traffic management. This way we can better plan for the future – e.g. modelling the impact on future air traffic and evolving scenarios, which will be critical to resilience and recovery,” says Dr Evelina Gabasova, a Principal Research Data Scientist at the Alan Turing Institute who leads the research engineering efforts in the project.  

Louisa Smith, Head of Research and Development at NATS said: “We are incredibly excited to be launching a new strategic partnership with the team at The Alan Turing Institute. We have only scratched the surface understanding the opportunities and limitations of AI within aviation. Working together means the future of air traffic control will have some of the brightest air traffic management and data science minds in the UK solving an enormously complex challenge – one that will ultimately equip our industry with the knowledge of how AI can aid a safer, more efficient and predictable air traffic management capability.” 

Europe