Air Traffic Control
Building a first-of-a-kind AI for Air Traffic Control, by blending advanced RL algorithms with “human-in-the-loop” safety considerations, to increase routing efficiency and reduce operator load.
Enhancing Transport Networks with ML
From route creation to traffic management and infrastructure development, our probabilistic approach to ML enables transport planners and operators to optimise complex networks safely.
Using ML to enhance multiple transport paradigms:
Digital Transformation
Balancing energy efficiency with safety and capacity are the defining challenges for transport networks. digiLab are experts in supporting organisations to apply ML across their systems, in order to fully leverage their data, optimise their activities, and deliver upon these challenges while minimising costs.
twinLab Platform
twinLab, our intuitive ML platform, enables analysts and engineers to augment the intelligence provided by sensors and other data sources. Through seamless integrations, twinLab brings the power of probabilistic ML to challenges like network optimisation or utilisation predictions
digiLab built a digital twin of the ATCO environment within three months by combining the best in modern software development practice, unique capabilities in probabilistic modelling, and a deep domain understanding of airspace management.
Dr Richard Cannon
R&D Lead
National Air Traffic Services
Get in touch to explore our technical demos:
Air Pollution - Digital Twin
Using probabilistic ML to understand the evolution of air pollution in urban environments, enabling predictive forecasting and improving the quality of response mechanisms.
How twinLab can help transport planners, developers and operators
- twinLab uses the latest in probabilistic modelling to build fast, robust emulators with built-in uncertainty quantification (UQ)
- Digital twins can be used to predict future states from limited information, or identify areas for more research
- Round the clock technical support, training & workshops, and extensive domain specific demos & tutorials
