Skip to content

AI For Aviation

Artificial intelligence applied to air traffic management can pave the way for new levels of predictability, safety and efficiency. Our team has been collaborating for several years now with EUROCONTROL and WaPT to safely reduce wake separation between flight. They recently published two papers :

The first one was co-written by our engineer Lorenzo and was presented a few weeks ago at the ICAS 2022 congress in Stockholm. His paper focuses on analysing the behaviour of the wakes’ turbulence. As incoming aircraft must wait until it is certain that the wake of the preceding aircraft will not disturb its landing, wakes represent one of the main origins of the time separation between planes. Eurocontrol and Euranova used data coming from LIDAR and sensors at the airports to predict the movement and the decay of the wakes and safely reduce the time separation.

Lorenzo says: “The conference was a great opportunity to meet and interact with the main actors of the aviation industry! We received feedback from other people working on wake separation and got an overview of how machine learning can be applied to the industry.”

The second is co-written by our engineer Luca and will be presented next week at EASN 2022 in Barcelona. He focused on departing planes and worked on optimising the predictions of aircraft behaviour (such as the trajectory or the speed profile) and wind to decrease the time between two consecutive departures while fulfilling all the required spacing and separation constraints.

Congratulations to the team!

Releated Posts

Augment to Interpret: Unsupervised and Inherently Interpretable Graph Embeddings

In this paper, we study graph representation learning and show that data augmentation that preserves semantics can be learned and used to produce interpretations. Our framework, which we named INGENIOUS, creates inherently interpretable embeddings and eliminates the need for costly additional post-hoc analysis.
Read More

SANGEA: Scalable and Attributed Network Generation

In this paper, we present SANGEA, a sizeable synthetic graph generation framework that extends the applicability of any SGG to large graphs. By first splitting the large graph into communities, SANGEA trains one SGG per community, then links the community graphs back together to create a synthetic large graph.
Read More