In the dark, crushing silence of the deep ocean, a new kind of sonar could soon be making waves—not with high-powered acoustic blasts or flashy emissions but with the barely perceptible hum of a stealth underwater drone’s propeller.
In a breakthrough that could revolutionize underwater exploration and military surveillance, research funded by the U.S. Navy demonstrated that autonomous underwater vehicles (AUVs) can use their self-generated noise to “see” beneath the ocean floor.
This novel approach, called through-the-sensor (TTS) sub-bottom imaging, eliminates the need for bulky and power-hungry sonar sources typically used in undersea mapping.
Instead, this new approach uses the low-frequency self-noise generated by an AUV’s propeller as a “source of opportunity” to reveal buried sediment layers, seafloor structures, and subsurface objects. No sonar pings. No external emitters. Just the soft, mechanical thrum of an underwater drone on patrol—turned into a stealthy acoustic eye.
The research, recently published in JASA Express Letters, opens the door to stealthier, more energy-efficient, and potentially more autonomous operations beneath the waves.
Researchers demonstrated their proof-of-concept during field tests at the New England shelf break, a region known for its complex undersea topography. A modified REMUS 600 AUV, outfitted with a short horizontal hydrophone array, gathered data as it cruised about 155 feet (35 meters) above the seabed.
The results showed that the passive images generated using the self-noise method closely matched those produced by the AUV’s active sonar system and reference scans conducted by the R/V Neil Armstrong using a Kongsberg multibeam profiler. Despite operating at lower frequencies and with less power, the passive method successfully identified key geological features, including a deep basement layer buried approximately 105 feet ( 32 meters) beneath the seafloor.
“A potential advantage of this TTS method, besides being totally passive, is that it provides a means for performing low-frequency sub-bottom imaging using an AUV,” researchers wrote. “Hence, this could potentially enable geoacoustic surveys with larger depth penetration, without the need to instrument the AUV with a separate active low-frequency source, which is typically associated with unique engineering challenges and payload requirements.”
A New Kind of Sonar
Sonar, or sound navigation and ranging, traditionally involves emitting sound waves into the water and recording their echoes to build images of the seafloor or underwater objects.
While effective, this approach has one major drawback: it’s loud. Active sonar can quickly reveal the presence of the vessel emitting it—a problem for submarines or AUVs operating in contested or clandestine environments.
However, this new method flips that model on its head. Instead of actively generating sonar pulses, the AUV uses its own mechanical sounds—specifically the low-frequency noise from its propeller—as a passive source of acoustic energy. That noise, typically considered a nuisance, becomes a tool for sensing the environment.
The technique, called Through-the-Sensor (TTS) sub-bottom imaging, involves towing a short horizontal line array of hydrophones behind the AUV. As the AUV moves along, its propeller emits a consistent stream of low-frequency sound.
As the sound travels downward, it reflects off the seabed and layers beneath it and is picked up by the trailing hydrophones. Using advanced signal processing algorithms—specifically, a variation of ray-based blind deconvolution—the system can infer the structure of the ocean floor and even what lies beneath it, like buried sediments or rock layers.
To make this work, the team focused on a frequency band between 100 Hz and 1 kHz. This range is characteristic of propeller noise and well-suited for penetrating below the seabed.
“This method is derived from the RBD [ray-based blind deconvolution] algorithm and thus potentially alleviates the need to instrument the AUV platform with a separate low-frequency active source,” researchers explained. “Additionally, the AUV’s propeller effectively acts as a compact low-frequency source, which indeed enables the use of the RBD algorithm.”
Why It Matters for Future Stealth Underwater Drones
The implications are significant from a defense perspective. Submarine warfare, undersea surveillance, and reconnaissance missions depend heavily on staying hidden.
While powerful, active sonar risks exposing the vehicle’s location to enemy sensors. Passive sonar systems—those that listen without transmitting—offer stealth but have traditionally required large, sensitive arrays and often rely on environmental noise or external signals.
What makes this TTS approach unique is its ability to transform a drone’s unavoidable emissions into a powerful imaging tool—repurposing the vehicle’s own acoustic signature without adding any new noise.
That dual-use aspect—converting waste into a sensor—makes the system stealthy and potentially more efficient. Without bulky or power-hungry active sonar systems, stealth underwater drones can carry less hardware, save energy, and extend mission duration.
Moreover, the hardware requirements are refreshingly simple. All that’s needed is an AUV with a towed hydrophone array and a processor capable of running the RBD algorithm. This significantly reduces the vehicle’s payload burden and simplifies deployment in tight or unpredictable environments.
For military applications, this means smaller, more autonomous stealth underwater drones capable of quietly mapping the seabed, identifying mines, or detecting submarine cables while remaining undetected.
The REMUS 600 tests conducted by researchers demonstrated that even though the passive images had lower resolution than those from an active sonar system, they were still accurate enough to reveal critical details like the shape of the seafloor and the presence of deeper, more reflective layers—useful markers in tactical or scientific operations.
From Ocean Science to Tactical Advantage
While the defense benefits are clear, the technique also has promising civilian and scientific applications.
Oceanographers, geologists, and environmental researchers often require sub-bottom imaging to understand sediment layers, tectonic structures, or undersea resources. Traditional methods require either large ships towing active sonar or long vertical arrays, both of which are costly and limited in maneuverability.
This new TTS method sidesteps many of those issues. Because the AUV can operate independently and passively, it could be deployed in sensitive environments where noise pollution or disturbance of marine life must be minimized. It could also help reduce the logistical burden of deep-sea surveys, making them more accessible to researchers and smaller institutions.
Notably, this approach aligns with a growing trend in robotics and autonomous systems: designing smarter machines that efficiently use their own data and byproducts. Just as some spacecraft now recycle waste heat for energy, and animals like bats and dolphins use bioacoustics to navigate, this technique embodies a biomimetic philosophy—treating the machine not merely as a tool but as a self-contained ecosystem of signals.
What’s Next: Toward Smarter, Stealthier Underwater Drones
Despite its promise, the current technique has limitations, and researchers note that improvements are needed before it becomes operational.
Chief among the challenges is that the self-noise signal is inherently weaker than active sonar pulses, requiring more extended data collection periods to build a clear image. Balancing the AUV’s movement, array stability, and environmental variability becomes crucial.
During testing, the team conducted a sensitivity study showing how shorter averaging times (2 seconds) yield lower-resolution images, while longer times (24 seconds) can wash out important spatial details if the terrain is highly variable. This could be a limitation in dynamic environments or faster-moving missions.
However, future work may address this by taking advantage of the full bandwidth of the AUV’s self-noise, potentially expanding beyond the 100 Hz to 1 kHz range. Better understanding the vehicle’s acoustic “fingerprint”—how its noise radiates at different angles and frequencies—could also help improve resolution and reduce artifacts.
Another frontier involves enhancing the signal processing techniques themselves. With more powerful onboard computing or cloud-based post-processing, future AUVs could adaptively tune their imaging strategies in real time, steering their listening arrays or changing course to optimize data collection.
Despite these challenges, for the U.S. Navy, this breakthrough represents a tantalizing step toward fully autonomous stealth underwater drones that can explore the depths in near silence, uncovering vital information all while remaining hidden from view.
Tim McMillan is a retired law enforcement executive, investigative reporter and co-founder of The Debrief. His writing typically focuses on defense, national security, the Intelligence Community and topics related to psychology. You can follow Tim on Twitter: @LtTimMcMillan. Tim can be reached by email: tim@thedebrief.org or through encrypted email: LtTimMcMillan@protonmail.com