Alan Turing Institute to spend UK.gov grants on AI for air traffic control and banking

Turns out the skies can be the limit for machine learning

The UK government will bankroll two major research projects led by the country's national institute for AI, the Alan Turing Institute: one to automate air traffic control, and the other for banking services.

Project Bluebird, arguably the more interesting of the two, will receive £11.9m and explore how techniques like reinforcement learning can be applied to coordinate the flight paths of aircraft. With fresh investment, the goal is to eventually conduct the world’s first live trials of an AI system capable of monitoring air traffic and communicating with pilots.

Researchers at the Alan Turing Institute have been working on this problem with NATS, Blighty’s biggest air traffic control provider, since 2018. It's hoped the machine-learning system can assist human controllers in preventing accidents and improving traffic flow. By optimizing journeys, the software may even reduce fuel consumption, too.

We're told the project will have three parts:

• Develop a probabilistic digital twin of UK airspace: this real-time, physics-based computer model will predict future flight trajectories and their likelihoods – essential information for decision-making. It will be trained on a NATS dataset of at least 10 million flight records, and will take into account the many uncertainties in ATC, such as weather, or aircraft performance.
• Build a machine learning system that collaborates with humans to control UK airspace: unlike current human-centric approaches, this system will simultaneously focus on both the immediate, high-risk detection of potential aircraft conflicts, and the lower risk strategic planning of the entire airspace, thus increasing the efficiency of ATC decision-making. To achieve this, researchers will develop algorithms that use the latest machine learning techniques, such as reinforcement learning, to optimise aircraft paths.
• Design methods and tools that promote safe, explainable and trustworthy use of AI in air traffic control systems: this will involve experiments with controllers to understand how they make decisions, so that these behaviours can be taught to AI systems.

The aim, as far as we can tell, is not to outright replace human air traffic controllers, and instead get humans and intelligent machines working closer and better together.

“The project will explore ethical questions such as where the responsibility lies if a human-AI system makes a mistake, how to build a system that is trusted by humans, and how to balance the need for both safety and efficiency,” the institute said.

Some of the code for this project is open-source and already up on GitHub.

AI bankers

The second funded scheme is a £12.4m collaboration between the institute and HSBC dubbed Project FAIR. An interdisciplinary team made up of computer scientists and economists will investigate how AI can revamp financial services and tools, and look for the best ways to apply the technology fairly and securely.

Interestingly, the project will involve generating synthetic financial data for researchers to experiment with. Using real data is risky, and isn’t possible for privacy reasons.

“It aims to address challenges such as how to increase the accuracy of predictive models without compromising privacy and security of customers, or improving transparency without leaving systems open to external threats,” the project’s description read.

“The project intends to outline how the finance sector can make the transformational shift to the greater use of AI technologies, ensuring they have fairness, security and accountability at their heart, while being robust and aware of privacy.”

The funding comes from the Engineering and Physical Sciences Research Council, part of UK Research and Innovation, a government body working under the Department for Business, Energy and Industrial Strategy. ®