The shift from aquatic to terrestrial life is considered one of the key milestones in Earth’s biological history. Currently, a collaboration of roboticists, paleontologists, and biologists is utilizing robotics to investigate how the ancestors of contemporary land animals evolved from swimming to walking around 390 million years ago.
The shift from aquatic to terrestrial life is considered one of the key milestones in Earth’s biological history. Currently, a collaboration of roboticists, paleontologists, and biologists is utilizing robotics to investigate how the ancestors of contemporary land animals evolved from swimming to walking around 390 million years ago.
In a study published in the journal Science Robotics, the research group led by the University of Cambridge details how ‘palaeo-inspired robotics’ can offer a valuable method for examining how the pectoral and pelvic fins of ancient fish transformed to support their weight on land.
“Due to limited fossil evidence, our understanding of how ancient life transitioned onto land is incomplete,” explained Dr. Michael Ishida from Cambridge’s Department of Engineering, the leading author of the study. “Paleontologists analyze old fossils for insights into the structure of hip and pelvic joints, but there are inherent limitations to what can be learned from fossils alone. This is where robots prove useful, as they can help fill gaps in our understanding, especially regarding major changes in vertebrate locomotion.”
Ishida is part of Cambridge’s Bio-Inspired Robotics Laboratory, which is directed by Professor Fumiya Iida, the senior author of the paper. The team focuses on creating energy-efficient robots for various uses, drawing inspiration from the ways animals and humans move efficiently.
With support from the Human Frontier Science Program, the team is working on developing palaeo-inspired robots, drawing on the movements of contemporary ‘walking fish’ such as mudskippers, as well as fossils of extinct fish. “In a lab setting, we can’t alter the walking patterns of live fish, and we certainly can’t animate a fossil, so we utilize robots to replicate their anatomy and behaviors,” said Ishida.
The team plans to create robotic versions of ancient fish skeletons, featuring mechanical joints that replicate the functions of muscles and ligaments. Once these robots are operational, they will conduct experiments to explore how these ancient creatures might have moved.
“We aim to discover details like the energy requirements of various walking styles, or which movements were most efficient,” noted Ishida. “This information can help validate or challenge existing theories about the evolution of these early creatures.”
A significant obstacle in this research is the scarcity of complete fossil records. Many ancient species from this period are only represented by fragmentary skeletons, complicating the effort to reconstruct their full range of motion.
“Sometimes, we’re left to speculate about how certain bones interacted or functioned,” Ishida remarked. “This is why robots are advantageous — they help verify these speculations and provide new evidence that can support or dispute them.”
While robotics is frequently used to investigate the movements of living animals, very few research teams apply this technology to extinct species. “Only a handful of groups are engaged in this type of research,” Ishida stated. “We believe this approach is a natural fit — robots can yield insights into ancient animals that fossils or contemporary species alone cannot reveal.”
The research team hopes to inspire other scientists to harness the potential of robotics to explore the biomechanics of long-gone creatures. “We strive to connect fossil records with real-world mechanics,” Ishida mentioned. “While computer models hold significant importance in this area of study, robots interact with the real world, allowing us to test theories about how these organisms moved and possibly even the reasons behind their movement patterns.”
Currently, the project is in its initial phases of building the palaeo-robots, with expectations to yield results within the coming year. The researchers believe their robotic models will not only enhance the understanding of evolutionary biology but also pave the way for new collaborations between engineers and experts from various disciplines.
The study received backing from the Human Frontier Science Program. Fumiya Iida is a Fellow of Corpus Christi College, Cambridge, while Michael Ishida serves as a Postdoctoral Research Associate at Gonville and Caius College, Cambridge.
