Harbor seals hunt fish in pitch-black, turbid water without relying on vision or sonar. They do it with their whiskers. A growing body of research now shows that the distinctive wavy geometry of seal vibrissae acts as a finely tuned hydrodynamic antenna, picking up faint underwater wakes left by swimming prey. That biological trick has caught the attention of engineers building the next generation of underwater robot sensors, and recent lab results suggest the artificial versions are getting remarkably close to matching the real thing.
How Seal Whiskers Read the Water
The foundational insight dates back to behavioral experiments showing that harbor seal vibrissae operate as hydrodynamic sensors, not merely as touch receptors. Unlike the smooth whiskers found on California sea lions, harbor seal whiskers have an undulating, wavy cross-section that varies in diameter along their length. That shape turns out to be central to their sensing power.
Follow-up work using isolated vibrissae to measure vortex shedding helped clarify the mechanism. When a seal swims forward, water flowing past its face would normally cause its whiskers to vibrate on their own, creating noise that drowns out useful signals. The undulating geometry suppresses those self-generated vibrations by more than ten times compared with a smooth cylinder, according to analyses from the MIT-WHOI Joint Program. With that self-noise stripped away, the whiskers become exquisitely sensitive to external disturbances, such as the rotating vortex street left behind by a swimming herring.
Hydrodynamic tracking tests suggest that a seal can follow the trail of a fish at distances of well over 100 meters. That range is remarkable for a passive sensor that requires no emitted signal and no electrical power. The animal simply reads the water’s lingering signature of a fish that passed by minutes earlier, using its muzzle as a dense array of flow probes.
The Slalom Effect and Active Whisking
One of the more striking findings came from tank experiments at MIT, where researcher Heather Beem built a scaled-up artificial whisker and towed it through the wake of a cylinder. As it entered the wake, the flexible model began to vibrate strongly and trace out a slalom-like path, weaving between alternating vortices rather than being buffeted randomly. This characteristic slaloming response is now considered a leading explanation for how seals lock onto and follow a prey trail with such precision: the whisker effectively “rides” the wake, converting subtle pressure differences into rhythmic motion that the nervous system can interpret.
Crucially, the whiskers are not just passive receivers. Seals actively move them, much like rodents that whisk their facial vibrissae to explore their surroundings. Research from the Kottapalli group at the University of Groningen, reported in npj Flexible Electronics, showed that seals exhibit exceptional ability to detect underwater prey even in complete darkness by sweeping their whiskers forward and back. In controlled experiments, the team measured a signal-to-noise ratio of about 4.7 during the retraction phase, higher than when the whiskers were held rigid. That difference matters because it means the act of whisking itself boosts sensitivity, much like how a person lightly drags their fingertips across a surface to feel texture more clearly. Treating a vibrissa as a static rod, as many early engineering models did, misses a key part of the animal’s sensing strategy.
From Biology to Biomimetic Sensors
Engineers have spent the past decade translating these biological principles into hardware. The design challenge has two parts: replicate the undulating geometry that kills self-noise, and achieve sensitivity fine enough to detect the faint pressure signatures left by underwater objects. That combination is difficult, because the same flow that carries useful information also threatens to shake the sensor itself.
Recent work in Bioinspiration and Biomimetics reports that seal-inspired devices are closing the gap. A 2025 study showed that carefully tuned artificial whiskers can detect wakes at flow speeds as low as 0.5 millimeters per second, a threshold the authors describe as comparable to biological performance. The prototype used low-cost materials and simple fabrication steps, aiming for sensors that could be deployed in large arrays rather than as one-off laboratory curiosities.
Other comparative tests have confirmed that sensors mimicking the undulating whisker shape show higher sensitivity than smooth designs, reinforcing the biological rationale for the wavy geometry. To understand why, morphology studies archived by NOAA examined dozens of real vibrissae and identified a key dimensionless wavelength-to-diameter ratio that characterizes the pattern of thickness variations along the whisker. Getting that ratio right appears to determine how effectively a sensor suppresses its own vortex-induced vibrations while remaining responsive to external wakes. If the surface is too smooth, the device generates excessive self-noise; if the undulations are too exaggerated, the structure becomes fragile and less predictable in flow.
Designers also have to consider how the whisker is mounted and instrumented. Some groups use strain gauges at the base to measure bending, while others embed flexible piezoelectric materials along the shaft to convert motion directly into voltage. The choice affects bandwidth and durability, especially for long-term deployment on autonomous underwater vehicles that may experience months of biofouling and mechanical stress.
What This Means for Underwater Robots
Most conventional underwater sensing relies on active sonar, optical cameras, or large acoustic arrays. Each has limits: sonar emits detectable signals that can reveal a robot’s presence, cameras fail in murky or dark water, and acoustic systems struggle with cluttered environments such as kelp forests, harbor pilings, or rocky reefs. Whisker-inspired sensors offer a passive alternative that works precisely where those tools fall short.
By integrating an array of whisker-like elements along a vehicle’s hull or nose, engineers hope to give robots a kind of “hydrodynamic vision.” Instead of forming an image from light or sound, the system would map patterns of flow, turbulence, and wake structures. In principle, this could allow a gliding robot to detect another vehicle passing hundreds of meters away, follow the trail of a moving target, or sense obstacles hidden in silt-laden water where sonar echoes become confusing.
Because the sensors are passive, they also lend themselves to low-power missions. A glider or drifting probe could cruise silently, sampling flow with whisker arrays and waking higher-energy systems only when something interesting passes nearby. For defense applications, such sensors could help detect quiet submarines or unmanned vehicles without broadcasting pings. For ecology, they could allow long-term monitoring of fish migrations or plankton blooms without disturbing the animals being studied.
There are still hurdles. Real ocean environments are far more chaotic than lab flumes, with overlapping wakes, waves, and background turbulence. Distinguishing a specific target’s signature from that noise will require sophisticated signal processing and machine-learning models trained on large datasets. Mechanical robustness is another challenge: flexible structures that are sensitive enough to register tiny flow changes must also survive entanglement with debris, repeated impacts, and biofouling.
Even so, the trajectory is clear. As biomimetic designs converge on the geometry and dynamics of real seal vibrissae, underwater robots are gaining access to a sensory modality that evolution has honed for millions of years. Instead of blasting the ocean with sound or relying on scarce photons, future vehicles may simply read the water the way a harbor seal does, feeling the ghostly trails that every swimmer, from a tiny fish to a massive submarine, leaves behind.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
