A freshwater fish from the Amazon that can hover, dart sideways, and reverse direction without moving its body has been the blueprint for a new class of underwater robot – one that achieves the same agility with a fraction of the hardware that previous designs required.
Building an underwater robot that moves like a living animal has been one of the harder unsolved problems in robotics. Propeller-driven vehicles are efficient in straight lines but poor at tight maneuvering. Robotic systems that mimic fish motion have historically required arrays of individually controlled actuators – motors or artificial muscles that make them expensive, mechanically complex, and prone to failure.
A new design from the University of California San Diego reduces that complexity dramatically, using a propulsion strategy that evolution refined over millions of years in a fish most people have never heard of.
The knife fish, native to South American rivers, moves through water using a single undulating fin that runs along the length of its belly. The fin generates thrust by propagating a continuous wave from one end to the other — forward for forward motion, reverse for backward, asymmetrically for turning — all without moving its body.
Engineers have known about the ribbon fin for decades. Copying it has been the challenge. The wave propagation that makes the fin effective requires coordinated motion along its full length, which typically means many independently controlled segments, each with its own actuator. The more actuators, the more wiring, more control complexity, more potential failure points, and more cost.
What the Researchers Built
The UCSD team built a soft ribbon fin from flexible silicone and drove it with far fewer actuators than previous designs — what engineers call a highly underactuated system. Underactuated means the robot has fewer motors than it has degrees of freedom.
A human hand, for comparison, has 27 degrees of freedom but only needs a fraction of that many muscles to perform most grasping tasks because of the way mechanical coupling distributes force.
The ribbon fin works similarly: a small number of actuators drive the fin at specific points, and the compliance of the soft material propagates the wave naturally through the rest of the structure.
The result is a robot that achieves agile swimming, including rapid turns and direction changes, with a mechanical design that is significantly simpler than comparable soft swimmers. The paper, published in IEEE Robotics and Automation Letters, confirms both the locomotion performance and the reduced actuator count.
What This Actually Changes
The bottleneck in underwater robotics has not been ambition. It has been cost and durability. Propeller-based autonomous underwater vehicles (AUVs) — the kind used for pipeline inspection, coral reef monitoring, and seafloor mapping — are effective but mechanically rigid and acoustically noisy. The propeller wash disturbs the environment around them, which is a real problem for scientific work. They also struggle in confined spaces and near delicate structures.
Soft underwater robots have been proposed as a solution to all of these problems. They are quieter, gentler, and more maneuverable. The obstacle has been that soft robotic propulsion systems have required too many components to be practical outside a research lab. Every additional actuator is a potential failure point in a saltwater environment where maintenance is difficult or impossible.
A ribbon fin design that achieves agile propulsion with minimal actuators changes that equation. Fewer actuators means fewer failure modes, simpler electronics, lower manufacturing cost, and a system that is easier to scale down to sizes where propeller-driven robots cannot operate. A small, soft, agile robot that can navigate around coral or inside a flooded pipe is a different tool from anything currently available in the field.
The Broader Context
This sits inside a wider push to move underwater robotics from specialised research equipment toward deployable field tools. Climate monitoring, offshore energy infrastructure inspection, and marine conservation all have growing demand for persistent, low-cost underwater sensing that does not disturb what it is observing. The ribbon fin approach is well matched to those requirements in a way that current propeller-driven AUVs are not.
What to Watch For
The current demonstration is a research prototype. Scaling soft ribbon fin propulsion to a fully autonomous, untethered vehicle that carries its own power, computing, and sensors adds engineering challenges the paper does not yet address. Battery integration, waterproofing at depth, and real-world navigation in currents all remain open problems.
The test that matters next is whether the design performs outside a controlled tank. Agility demonstrated in still water does not automatically transfer to the turbulent, variable conditions of a real ocean, river, or flooded infrastructure environment. That gap between lab performance and field performance is where most soft underwater robots have stalled before.
Sources
Adibnazari, I., Jeong, S. and Tolley, M.T. “Bioinspired Agile Swimming with a Highly Underactuated Soft Ribbon Fin.” IEEE Robotics and Automation Letters, Vol. 11, No. 5, pp. 5502-5509 (May 2026). DOI: 10.1109/LRA.2026.3673909 URL: https://ieeexplore.ieee.org/abstract/document/11434505
More Like This
MIT’s Inchworm Robots Just Changed the Math on Construction
Sam Brown holds a BSc in Electronic Systems Engineering from University of Regina, a Masters from the University of Alberta and a PhD from the University of Toronto in Electrical Engineering. His reporting interests include Physics and Artificial Intelligence.