What looks like a fish, swims like a fish but isn’t a fish? The latest in soft-bodied robots created by team of engineers of the Computer Science and Artificial Intelligence Laboratory at the Massachusetts Institute of Technology.

The team, comprised of Daniela Rus, Professor of Electrical Engineering and Computer Science and Director of the Computer Science, Cagdas Onal, Assistant Professor of Mechanical Engineering at the Worcester Polytechnic Institute, and Andrew Marchese, a doctoral candidate in engineering at MIT, created the robot to be autonomous, this means it has all the necessary components on board. Its flexible body is made of silicone rubber. The robot fish is able to swish from side to side as onboard gas inflated and deflates different parts of its body. This means that the robot is flexible and can maneuver as dexterously as a real fish—making a full C-turn with its body in just 100 milliseconds.

According to Rus, this type of soft robot is “inherently safe,” because unlike most robots, if this one crashed into you, it won’t crush your bones. “As robots penetrate the physical world and start interacting with people more and more, it’s much easier to make robots safe if their bodies are so wonderfully soft that there’s no danger if they whack you,” said Rus.

The fish can perform 20-30 agile escape maneuvers before its carbon dioxide canister runs out. The team is working on an updated version that will be able to swim for around 30 minutes using water to inflate its interior channels rather than carbon dioxide.

In the not so distant future, the fish-bot could be put to use for covert science missions where it might ab able to infiltrate schools of real fish to collate data about their behavior.  

The field of soft robotics holds great potential for the development of smart machines that can adjust their shape and size to fit variable environments and interact with living things without causing them harm. Possible applications include: mine detection, assistive healthcare, search and rescue missions and space instrument repair.


TechEmergence conducts direct interviews and consensus analysis with leading experts in machine learning and artificial intelligence. Stay ahead with of the industry with charts, figures, and insights from our unparalleled network, including executives from Facebook, Google, Baidu, Yahoo!, MIT, Stanford and beyond: