AI, a drone and the 5G network allow real-time detection of foreign objects on the runway.

Pioneer project at San Sebastian airport: Artificial Intelligence, a drone and the 5G network allow real-time detection of foreign objects on the runway.  

Aena, Spain's leading airport manager by passenger volume, has entrusted Inetum, a European leader in digital services and solutions, with this pioneering project in the world. The solution, which has been implemented as a proof of concept at San Sebastian Airport, allows to detect FODs (Foreing Objects Debris) on the runway in real time, using images sent by a drone through a 5G network, and significantly reduces visualization and reaction times.

A drone, equipped with sensors and cameras, captures and transmits track image in real time. If it detects a foreign object, which may pose a safety risk for landing and take-off operations, a warning is triggered and the object is geolocated, so that it can be removed. The solution developed by the technology company Inetum uses an AI-based algorithm that analyzes images in real time, transmitting them through 5G connectivity, which is already a reality at San Sebastian airport, where Aena and Cellnex Telecom have installed the first private network based on 5G technology in Spanish airports and one of the first in the European airport landscape.

This real-time FOD detection project has already passed several scenarios and functional tests. The algorithm allows effectively detecting and discriminating FODs in real time through the images transmitted by the drone.

"Aena required a custom model”, says Adrián Gómez Pastor, director of Industry & Security BU at Inetum Solutions in Spain, "starting from an existing AI engine as a base, to feed it with high quality recorded images and train it (machine learning) until the minimum quality requirements were met. In addition, the model must be exportable to any standard analytics platform and have all the necessary cybersecurity mechanisms to ensure the inviolability of the information generated. And the algorithm has to work with images collected in different operating conditions: day, night, rain, fog..."

The Inetum team has developed an algorithm based on Artificial Intelligence that allows to visualize and store the images transmitted in real time and their processing, both in real time and through historical records; to store the data obtained from the processing to visualize or extract them later, according to a series of parameters (detection time, exact location in altitude and latitude coordinates, dimensions of the FOD, primary image captured from the FOD). Once the object is detected, an alarm is displayed on the interface that warns in real time of the existence of the FOD and its persistence, so that its criticality can be perceived.

The term FOD includes a wide range of materials, including loose pieces, fragments of pavement, construction materials, pieces of luggage, or remains of fauna. With this project, the use of drones, we are able to apply technology to continue to maintain the safety levels we have on airport runways.

To ensure the success of this project, Inetum has counted with the entrepreneurs of Invicsa Airtech, a startup from Leon that offers design, optimization and operation with drones and that are a reference in security and aerial operations with unmanned aircraft. Vicomtech, an applied research technology center specialized in AI, Visual Computing & Interaction, has also collaborated with Inetum, which has its own integral management methodology that guarantees the optimization of processes and the transfer of technology to companies in a transparent manner.

Inetum brings its experience and knowledge to support the transformation of the aerospace industry, with new services and digitization projects that reinforce safety and bring a positive impact to society.