Contact form
Your request has been sent.
Please enter your name
Please enter your email
Please enter a correct email address
Please enter your company name
Please enter your message

A concrete AI application: Multi sources data analysis to localize planes

← All resources
Technical Takes
A concrete AI application: Multi sources data analysis to localize planes

On May 18, 2020, an allied frigate operating in the Syrian channel detects two MIG-29s taking off from the Russian-controlled military base of Latakia in Syria.

Since the fall of Colonel Kadhafi in 2011, the country has been in chaos. Two powers are facing each other: the Government of National Accord (GNA) of Fayez al-Sarraj (supported by Turkey and Qatar, and recognized by the UN) and Marshal Haftar, head of the Libyan National Army (LNA), who is challenging the power and waging a war (supported by Russia in particular). France is closely following the evolution of this conflict.

In this context, how can artificial intelligence help decision makers in the analysis of the situation?

When the two MIGs pass by, our naval force headquarter is immediately notified and the French political authorities give the order to locate them.

The intelligence teams then take over and start the intelligence process.

To help them, they use the "multi-intelligence" Preligens tool which, thanks to artificial intelligence, aggregates data (commercial and open source) from multiple sensors (satellite images, air and ship transponder data, infrared imagery, social networks, etc.) and alerts analysts in case of abnormal changes in the situation at a given theater of operations.

The tool allows them to consult different data streams to find the MIG-29s

They start with the very high resolution images available on Latakia, the base from which the planes are likely to have taken off. Thanks to AI algorithms, the MIG-29s are indeed located on the airbase, and the intelligence professionals can testify of another change: the planes have been repainted in the colors of Haftar's army.

By then consulting the Twitter feeds that are mentionning MIG-29 around the 18th of May, they quickly learn that these planes would have landed at Al Jufra airport, which they will then seek to confirm.


Night light

They then consult another image stream called "Night light" and note some activity at Al Jufra airport on the night of the 18th to 19th of May. No activity had been detected before this date, which may confirm an abnormal situation.

Al Jufra

From that moment, the intelligence agents only need to try and find traces of these MIGs at Al Jufra airport on May 19, 2020. 

Thanks to the object detection algorithms that continuously and automatically run on satellite images, they manage to locate a MIG 29 towed on the runway.

At this stage, the information is perfectly qualified and the Chief of Staff is in a position to make the necessary decisions, such as whether or not to redirect its sovereign sensors.