SOFI – MIT’s Robotic Fish

Sofi Blog Image

A team of computer scientists from MIT have developed a soft robotic fish that can swim on its own alongside real fish in ocean environments. The journal Science Robotics published an article today documenting the robot fish.

Researchers from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) announced “SoFi,” a robotic fish that can swim at depths of more than 50 feet, with the ability to swim in a straight line, turn or dive up and down. The fish has an undulating tail and the ability to control its own buoyancy, researchers said. It can be operated by a waterproof Super Nintendo game controller.

“To our knowledge, this is the first robotic fish that can swim untethered in three dimensions for extended periods of time,” said Robert Katzschmann, CSAIL Ph.D. candidate and lead author of the journal article. “We are excited about the possibility of being able to use a system like this to get closer to marine life than humans can get on their own.”

Other autonomous underwater vehicles (AUVs) typically are tethered to boats or use bulkier propellers, the researchers said. On SoFi, the lightweight setup includes on-board sensors for perception, a servo motor and lithium polymer battery similar to ones found in smartphones.

The goal of SoFi is to be as non-disruptive as possible, with minimal noise from the motor to low-frequency emissions from the communication systems of the controller. Commands are sent at wavelengths of 30 to 36 kHz.

“The robot is capable of close observations and interactions with marine life and appears to not be disturbing to real fish,” said Daniela Rus, CSAIL director, who also wrote the paper with Katzschmann, graduate student Joseph DelPreto, and Robert MacCurdy, an assistant professor at the University of Colorado at Boulder.

Next steps for the team include working on additional improvements, including increasing SoFi’s speed, and to build additional models for biologists to “study how fish respond to different changes in their environment, the university team said.

“We view SoFi as a first step toward developing almost an underwater observatory of sorts,” Rus said. For example, the article said the SoFi prototype “provides the opportunity to perform studies of the biocenosis [organism association] of coral reefs and other marine environments within natural habitats.”

Researchers could also use the prototype to “easily change its size, color, and shape to emulate various types of fish with different dynamic behaviors.” Because of its small size, the soft fish could be rapidly fabricated to create a “swarm of robotic fish,” allowing for the studies of schools of fish and interactions with different ocean dynamics, the paper said.

Possible commercial applications for the system include using several of the fish to create a network of sensor nodes that can swim and record data, said Katzschmann.

“The fish can not only gather video, but potentially also other sensor data, as well as taking water samples,” Katzschmann said. In addition, the system could be used for underwater inspection of infrastructure such as gas lines or oil rigs.