Researchers have developed new machine learning tools to help ease air traffic controllers’ workload.
The MALORCA (Machine Learning of Speech Recognition Models for Controller Assistance) project was funded within the framework of the SESAR Joint Undertaking, a public-private partnership set up to modernise Europe’s air traffic management system.
As the MALORCA project website explains, air traffic control (ATC) instructions are usually given via voice communication to pilots. But ATC systems, to be safe and efficient, need up-to-date data. This requires lots of inputs from air traffic controllers – mostly using mouse and keyboard – to keep the system data correct.
Automatic speech recognition, which converts speech to text, is an alternative that can reduce controllers’ workload and increase the efficiency of air traffic management (ATM) – resulting in fuel savings of 50 to 65 litres per flight.
“Fortunately, automatic speech recognition has reached a level of reliability that is sufficient for implementation into an ATM system,” said project coordinator Hartmut Helmke. “However, we need to reduce the transfer costs of speech recognition systems from one approach area to an other one.”
Currently, assistant-based speech recognition requires manual adaptation to the local environment (e.g. regional accents, phraseology deviations and local constraints).
The MALORCA project aimed to automate this re-learning, adaptation and customisation process by automatically learning local speech recognition and controllers’ models from radar and speech data recordings.
Machine learning enables computer systems to ‘learn’ and improve their performance on specific tasks over time, without being explicitly programmed.
The new machine learning tools will replace much of the manual effort previously required, making adaptation to different airports and maintenance cheaper and faster.