For years, science fiction has promised a future filled with robots that can swim, crawl and fly like animals. In one research lab at Caltech, what once felt like distant imagination is becoming reality.

At first glance, they look like any other moon jellyfish — soft-bodied, translucent and ghostlike, as their bell-shaped bodies pulse gently through the water. But look closer, and you’ll spot a tangle of wires, a flash of orange plastic, and a sudden intentional movement.

These are no ordinary jellies. These are cyborgs.

At the Dabiri Lab at Caltech, which focuses on the study of fluid dynamics and bio-inspired engineering, researchers are embedding microelectric controllers into jellyfish, creating “biohybrid” devices. The plan: Dispatch these remotely controlled jellyfish robots to collect environmental data at a fraction of the cost of conventional underwater robots — and potentially redefine how we monitor the ocean.

Caltech graduate student, Noa Yoder, a graduate student is one of the members of a team developing cyborg jellyfish, using a device made up of a container filled with sensors and two electrodes that attach to the jellyfish’s muscles.

Advertisement

“It fills the niche between high-tech underwater robots and just attaching sensors to animals, which you have little control over where they go,” said Noa Yoder, a graduate student in the Dabiri Lab. “These devices are very low cost and it would be easy to scale them to a whole swarm of jellyfish, which already exist in nature.”

Jellyfish are perfect for the job in other ways: Unlike most marine animals, jellyfish have no central nervous system and no pain receptors, making them ideal candidates for cybernetic augmentation. They also exhibit remarkable regenerative abilities, capable of regrowing lost body parts and reverting to earlier life stages in response to injury or stress — they can heal in as quickly as 24 hours after the removal of a device.

The project began nearly a decade ago with Nicole Xu, a former graduate student in the Dabiri Lab who is now a professor at the University of Colorado Boulder. Her research has demonstrated that electrodes embedded into jellyfish can reliably trigger muscle contractions, setting the stage for field tests and early demonstrations of the device. Later, another Caltech graduate student, Simon Anuszczyk, showed that adding robotic attachments to jellyfish — sometimes even larger than the jellyfish themselves — did not necessarily impair the animals’ swimming. And in fact, if designed correctly, the attachments can even improve their speed and mobility.

The device allows the scientists to pilot the jellyfish through ocean waters and record measurements of pH, temperature, salinity, etc.

Advertisement

There have been several iterations of the robotic component, but the general concept of each is the same: a central unit which houses the sensors used to collect information, and two electrodes attached via wires. “We attach [the device] and send electric signals to electrodes embedded in the jellyfish,” Yoder explained. “When that signal is sent, the muscle contracts and the jellyfish swims.” By triggering these contractions in a controlled pattern, researchers are able to influence how and when the jellyfish move — allowing them to navigate through the water and collect data in specific locations.

These cyborg devices offer a cheaper (and more sustainable) option for marine research, making it more widely accessible for anyone looking to study the ocean.

The team hopes this project will make ocean research more accessible — not just for elite institutions with multimillion-dollar submersibles, but for smaller labs and conservation groups as well. Because the devices are relatively low-cost and scalable, they could open the door for more frequent, distributed data collection across the globe.

The latest version of the device includes a microcontroller, which sends the signal to stimulate swimming, along with a pressure sensor, a temperature sensor and an SD card to log data. All of which fits inside a watertight 3D-printed structure about the size of a half dollar. A magnet and external ballast keep it neutrally buoyant, allowing the jellyfish to swim freely, and properly oriented.

One limitation of the latest iteration is that the jellyfish can only be controlled to move up and down, as the device is weighted to maintain a fixed vertical orientation, and the system lacks any mechanism to steer horizontally. One of Yoder’s current projects seeks to address this challenge by implementing an internal servo arm, a small motorized lever that shifts the internal weight of the device, enabling directional movement and mid-swim rotations.

Unlike most marine animals, jellyfish have no central nervous system and no pain receptors, making them ideal candidates for cybernetic augmentation.

But perhaps the biggest limitation is the integrity of the device itself. “Jellyfish exist in pretty much every ocean already, at every depth, in every environment,” said Yoder.

To take full advantage of that natural range, however, the technology needs to catch up. While jellyfish can swim under crushing deep-sea pressures of up to 400 bar — approximately the same as having 15 African elephants sit in the palm of your hand — the 3D-printed structures warp at such depths, which can compromise their performance. So, Yoder is working on a deep-sea version using pressurized glass spheres, the same kind used for deep-sea cabling and robotics.

Other current projects include studying the fluid dynamics of the jellyfish itself. “Jellyfish are very efficient swimmers,” Yoder said. “We wanted to see how having these biohybrid attachments affects that.” That could lead to augmentations that make future cyborg jellyfish even more useful for research.

The research team has also begun working with a wider range of jellyfish species, including upside-down Cassiopeia jellies found in the Florida Keys and box jellies native to Kona, Hawaii. The goal is to find native species that can be tapped for regional projects — using local animals minimizes ecological risk.

“This research approaches underwater robotics from a completely different angle,” Yoder said. “I’ve always been interested in biomechanics and robotics, and trying to have robots that imitate wildlife. This project just takes it a step further. Why build a robot that can’t quite capture the natural mechanisms when you can use the animal itself?”