France AI Submarine Identification: In an increasing sea of underwater sounds, the French Navy is using artificial intelligence to assist its submariners in identifying enemy vessels.
The chief of the Navy’s acoustic detection and interpretation centre, CIRA in Toulon, Vincent Magnan, stated in a presentation held here on Thursday that the institution is collaborating with a French startup called Preligens on AI-powered analysis of underwater acoustic signals. By the end of the year, France plans to test the technology aboard its submarines, with an operational deployment date set for 2025.
The amount of data being gathered for analysis is expanding dramatically as France outfits more and more vessels with potent passive acoustic sensors. To help its acoustics analysts—also known as “golden ears”—cut through the cacophony both at the Toulon centre and onboard its submarines, the Navy is relying on artificial intelligence (AI).
US Navy Places $290 Million Small Boat Order with Gravois Aluminium Boats
Magnan predicted that there would be “a massive flow of data” as a result of more sensors and longer detection ranges. “We need to turn to technological innovations, including artificial intelligence, to be able to analyze all this data, and especially to be able to isolate from it the useful and decisive information for the conduct of our combat operations.”
According to Magnan, in addition to aircraft, frigates, and submarines equipped with passive sensors, the near future will see the introduction of underwater gliders and drones that can record sound waves. According to Defense News, CIRA gathered one terabyte of data in 2020, roughly 10 terabytes by 2024, and 100 terabytes or more are predicted to be acquired by 2030.
Because it enables surface ships and submarines to detect underwater sounds during operations at sea and deduce tactical aspects in “all discretion,” without an adversary knowing about it, “passive acoustic warfare” is gaining popularity, according to Magnan. The Navy may be able to estimate a target’s speed through a certain propulsion pattern, which can subsequently be used to determine a tactical move.
The Toulon Center is using artificial intelligence (AI) to separate interesting acoustic data so that humans can carry out high-value additional analysis. At sea, the objective will essentially be the same, with AI freeing up human operators to concentrate on the valuable signals.
Therefore, Magnan explained, “So we use technology to discard or filter the standard part of the signal, the almost useless part, and we rely on humans to exploit the useful part,”
US Receives Anti-Drone Directed Energy Weapons from Epirus
According to Magnan, it takes two “golden ears” more than forty working days to go through the twelve days’ worth of acoustic data collected in the waters near Toulon. Preligens’ AI demonstration can extract relevant signals from those same recordings in 4-5 hours, whereas human analysis takes an extra 5–6 days. “So it’s clear that there has been a huge gain already.”
While CIRA examined five-minute audio recordings of a specific danger in the 1990s and 2000s, the centre now processes data spanning forty days, which, in the words of the centre’s head, “requires a great deal of human capacity” to process.
By 2020, a sonar operator could view over 200 kilometres and trace 100 tracks, compared to the early 2000s, when they could only see about 20 kilometres and monitor ten simultaneous acoustic interactions, according to Magnan. According to the commander, there will be a genuine need to simplify the detection duty because France’s third-generation ballistic missile submarines will have much stronger sensor capabilities.
Currently operating four nuclear-powered ballistic missile submarines of the Le Triomphant class, France is currently replacing its six nuclear-powered attack submarines of the Rubis class with Suffren-class vessels.
According to Magnan, the AI model has demonstrated “very encouraging results,” being able to differentiate between recreational and commercial boats as well as determine propeller speed, propulsion systems, and even the number of propeller blades. The integration of the AI models used for acoustics with data from satellite, radar, visual, and electromagnetic sources will be a future step.
According to Julian Le Deunf, an expert at the Armed Forces Ministry’s recently established agency for AI in defence, the team working on acoustic detection has developed a tool to automatically detect and identify various acoustic sources and sound emissions. The tool will be showcased at the Viva Technology show in Paris next week.
“The promising results over these last few months also encourage us to test all these capabilities in real-life conditions, so to take the jump on board the submarine and test these models directly at sea,” Le Deunf said. “The goal for the end of the year is really to succeed in plugging the model directly behind an audio stream, behind a sensor.”
Indian Army reduces ammo imports significantly with Make in India Initiative
After Magnan met with Preligens executives in October of 2021, the AI project began operating at CIRA. The company’s artificial intelligence (AI) products were already being used by French military intelligence to analyze satellite photos, and according to Magnan, his discussions with Preligens gave rise to the notion that the model could be duplicated to interpret signals from underwater.
According to Magnan, eventually, the AI algorithms will be able to distinguish background noises like a pump starting up or a wrench falling in a hold.
He stated, “The idea in the long run is obviously to find models that are effective and efficient over the whole acoustic spectrum of the sources we encounter at sea.”
