Skip to main content

Robotics ‘Revives’ a Long-Extinct Starfish Ancestor

Engineers and paleontologists teamed up to reconstruct an ancestor of starfish from the Paleozoic era and figure out how it moved

Cystoid (Pleurocystites filitextus) fossilized in limestone.

Cystoid (Pleurocystites filitextus) fossilized in limestone.

Despite incredible advances in modern genomic research, science is nowhere near being able to clone long-extinct animals like the fictional ones in Jurassic Park. Even relatively recent extinctions remain enormously difficult to overcome. An innovative branch of research that joins robotics with paleontology, however, does let scientists bring back long-gone creatures in a different way: not with cells and DNA but with engineering skill and batteries.

An interdisciplinary team has built a robotic mimic of a bizarre and extinct ancestor of modern starfish. Through the work, published last week in the Proceedings of the National Academy of Sciences USA, the researchers have created a window into how one branch of echinoderms (the animal clade that includes starfish, brittle stars, sea urchins, sand dollars and sea cucumbers) might have evolved and moved around the ancient ocean floor. And this robotic revival could also spur future innovations in engineering and design.

“For many reasons, Jurassic Park would be impossible to produce,” says Imran Rahman, a paleontologist who researches animal evolutionary origins at London’s Natural History Museum and was not involved in the new study. Instead this robot “is the closest we’re ever going to get to one of these animals alive.”


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


The animal in question is Pleurocystites, a genus of marine invertebrate that lived about 450 million years ago during the Paleozoic era and is thought to be one of the first groups of echinoderms capable of free motion. Pleurocystites was bilaterally symmetrical, as opposed to many of its relatives, which were radially symmetrical. It had a hard, calcified central body called a theca with three appendages: two short and curved feeding parts called brachioles on one end and one longer, muscular appendage called the stem on the other.

None

Image of Rhombot robot testbed inspired by anatomy of pleurocystitid. Credit: Department of Mechanical Engineering, Carnegie Mellon University

The small creatures, just centimeters in length, are common in the fossil record. Yet almost nothing is known about their life or locomotion, says Samuel Zamora, a paleontologist and echinoderm researcher at the Spanish Geological Survey and one of the new study’s authors. Rahman agrees: “For many years, I’ve been trying to figure out how these extinct weirdos were living. You know, how did they move or feed?” The new study, he adds, “is just a really, really exciting way of tackling one of these very long-standing questions.”

Engineers often take inspiration from nature. In 2017 researchers developed a synthetic material that can change textures to boost camouflage like octopus skin does. More than a decade before that, geckos began to inspire new adhesives. Scientists have also created so-called biomimetic robots, modeled after living creatures, to study animal behavior. One early example is from 1995, when Barbara Webb introduced a cricket bot intended to offer insights into cricket mating behavior. And this new Pleurocystites bot is not the first time that roboticists have re-created animals or their parts from fossil records to infer how once living creatures navigated the world. But the new study is the first to create a robot version of an extinct echinoderm, and it is also unique for its inclusion of paleontologists in the research team, says Talia Moore, a robotics researcher and mechanical engineer at the University of Michigan, who was not involved in the new study. “I think it’s a really beautiful blend of paleobiology and bioinspired robotics,” Moore says. “It’s rare to see such a deeply interwoven study between these two fields.”

The researchers created their robot designs based on an analysis of the fossil record by Zamora and another paleontologist and on computer modeling. They ran virtual simulations to test the feasibility of various movement hypotheses and then constructed scaled-up Pleurocystites mimics. The soft appendages were made of silicone and elastomers, and coils of shape-memory alloy enabled the robots’ “tail” to imitate the animal’s muscular stem. The scientists tested their bots in a fish tank on a surface that simulated firm ground covered with a layer of water.

Through different trials, the engineers changed various design elements—such as the stem appendage’s length, how stiff it was and how much it moved—to determine what sort of motion might have been most advantageous for navigating the Paleozoic seabed. In the process, the team went through “a depressing amount” of robots, says Richard Desatnik, co-lead author of the study and a mechanical engineering Ph.D. student at Carnegie Mellon University. In the end, the researchers found that side-to-side stem motion likely propelled Pleurocystites in a brachiole-forward direction. The researchers also homed in on an ideal stem length (about four times the length of the theca), gait (wide, sweeping motions) and stiffness (rigid rather than flexible).

The resulting robot model moved about as fast as expected, based on comparisons with similarly sized modern relatives. Its proportions closely matched the fossil record and provided a possible explanation for paleontological evidence that Pleurocystites’ stem evolved to be longer over time. The findings “make sense” and provide a likely answer to the enduring mystery of Pleurocystites’ locomotion, Rahman says.

Despite the robot’s slow and clumsy-looking wiggle, Moore says it also holds potential engineering lessons. Designing biologically inspired robots leads to more diverse and dynamic forms and “really pushes us to innovate in new ways,” she adds. In this instance, says study co-author Carmel Majidi, a mechanical engineer at Carnegie Mellon University, copying the extinct echinoderm prompted new ideas for combining soft and rigid robotic components.

Still, the Pleurocystites robot has limits. For one, “it’s impossible to know for sure” exactly how these extinct animals moved, Moore notes. The robot represents a good guess supported by a clever physics demonstration—but it’s obviously not definitive proof. And although built to scale, the robot was about four times larger than the fossilized organisms themselves. Majidi says he’d like to attempt a smaller version. Plus, he and his colleagues want to test the robot on different substrates. Pleurocystites are thought to have lived and moved across all manner of seafloor types—squishy, mucky, sandy and rocky—and different ground conditions could drastically alter what’s most advantageous for movement.

John Long, a biologist, cognitive scientist and co-founder of Vassar College’s Interdisciplinary Robotics Research Laboratory, points out that the study didn’t examine how the variables of stem length, stiffness and gait interacted—nor did the researchers test changes in the frequency of stem oscillation. “They’ve got a start,” Long says. Although “it’s a very important first step in studying these echinoderm fossils,” the study is not a complete exploration of all the Pleurocystites possibilities on its own.

Much remains to be uncovered about this long-lost invertebrate. But at least Jurassic Park fans can rest safe in the knowledge that neither the extinct animal nor its robotic clone is capable of opening doors.