US Navy tests AI-powered sonar to hunt submarines in RIMPAC drills

The U.S. Navy has tested an artificial intelligence-powered sonar system designed to improve submarine detection during the multinational Rim of the Pacific (RIMPAC) exercise near Hawaii.

The system, called SensorMAX and developed by Lockheed Martin, uses artificial intelligence and machine learning to analyse acoustic data collected by underwater sensors, identify potential targets and rapidly update its detection models, according to Military Times. 

During the July exercises, U.S. Navy MH-60R helicopters deployed sonobuoys that collected underwater acoustic signals and transmitted the data to SensorMAX. The system then assisted helicopter crews in identifying potential submarine targets.

“Ground operators retrained SensorMAX’s models on new sounds in minutes, then transmitted AI-enhanced undersea intelligence OTA [over the air] using an encrypted data link directly back to the helicopter and the fleet,” Lockheed Martin said in a news release.

The company described SensorMAX as the “Navy’s AI portable sonar expert.”

The technology is designed to create a continuous flow of information between airborne crews, ground operators and networks of underwater sensors. When a new acoustic signature is detected, operators can identify the sound, retrain the AI model and send an updated version back to aircraft in the field.

“Taking in subsurface acoustic data from a network of sensors, SensorMAX provides a continuous data stream to the aircrew and sends information to the ground station,” the company said. “When a new target is identified, system operators label the identified noise signature, retrain the model on the new data and push a small update packet back to the aircrew.”

According to Lockheed Martin, the system allows aircrews to monitor as many as eight sonobuoy surveillance feeds simultaneously, doubling the capacity of older systems. Each model retraining cycle can take less than five minutes, allowing operators to adapt to newly detected acoustic patterns during an ongoing mission.

The ability to rapidly retrain the system could be particularly important in anti-submarine warfare, where distinguishing an adversary's submarine from background ocean noise can be difficult. Submarine detection also depends on changing environmental conditions and the availability of accurate acoustic data.

SensorMAX can integrate information from different types of sensors and is designed for use across air, naval and ground operations.

“Lockheed Martin’s collaboration with FUSE Inc. enables secure, resilient software delivery to deployed aircraft, allowing SensorMAX’s trained AI models and mission software to be updated rapidly without interrupting operations,” the company said.

The system is part of a broader effort by the U.S. military to incorporate AI into naval warfare and accelerate the process of detecting and tracking underwater threats.

“By demonstrating platform-agnostic AI intelligence and classification, the future fleet can leverage any acoustic sensor to expand the protective ring around high-value assets and accelerating the ASW kill chain,” Lockheed Martin continued.

The development also reflects growing competition over underwater warfare technologies. China is pursuing similar applications of artificial intelligence for anti-submarine warfare. Chinese media reported last year that researchers had developed AI technology capable of assessing data from multiple sources, including sonobuoys and information on water temperature and salinity.

The growing use of AI in submarine detection could therefore reshape anti-submarine warfare by allowing forces to process larger volumes of sensor data, recognise acoustic patterns more rapidly and distribute updated intelligence to platforms operating at sea and in the air.

By Sabina Mammadli