Naro-nanin educational robot fish

The naro-nanin robot can be customized with different fins by connecting them to the alumi...

A new breed of robot fish that is both relatively inexpensive and highly customizable is teaching students between the ages of 10 and 18 about technology and biology. It’s the latest in a line of biologically-inspired underwater robots developed within the naro (nautical robots) project at ETH Zurich (Swiss Federal Institute of Technology), which has previously developed robots based on tuna fish and sea turtles.

Stefan Bertschi, who leads a school outreach program to promote young talent and raise awareness of the university’s programs, approached the naro team with the idea of an edutainment robot based on the tuna fish robot they had developed. Fast forward 4 months and the prototypes were swimming in a pool thanks to financial backing from SATW (the Swiss Academy of Engineering Sciences) and the naro team’s expertise.

Children aged 10 to 18 can experiment with different fin shapes, sizes, and placements to ...

“Our goal is to teach students different skills in the areas of biology, mathematics, physics, and computer science,” explains Cédric Siegenthaler, Project Leader of the naro (nautical robot) team at ETH Zurich. “They learn by playing and by combinations of scientific, technical, design, and artistic components. Depending on the age and level of the kids, naro-nanins can be configured with fins or fully open for kids that want to program the motion parameters themselves.”

The naro-nanin’s body consists of an anodized aluminum frame surrounding a polycarbonate tube. Inside, a plexiglass structure contains a Raspberry Pi running Linux and ROS as its main controller, a Li-Po battery that lasts about 2 hours, and a pump that fills a diving cylinder. The whole package weighs 7 kg (15 lb) and is 50 cm (19.6 in) long.

The kit comes with a set of fins, but students are encouraged to make their own. Different “species” can be created by changing the size and shape of the fins and placing them anywhere on the robot’s body. The fins are connected to four attachable actuator modules that contain two standard Hitec servo motors (for flapping and fin rotation). These, and playful decorations, can be freely positioned along the aluminum frames using Lego Technic connectors.

Students can experiment and see how fin size, shape, and placement on the robot’s body affect its speed and control. For example, they’ll quickly learn that if the robot’s weight isn’t balanced correctly it will have difficulty swimming properly. And they’ll learn about buoyancy and gravity, and how certain species of fish use a swimming bladder to dive and control depth. They can also hold competitions to see which design swims the fastest or is the most agile.

A pressure sensor detects the robot’s depth, while an inertial measurement unit senses and returns its pose to an Android app. A camera module is currently in the works that will allow you to see from its perspective, making underwater navigation easier. Currently the robot is remotely-controlled using a Bluetooth game pad, but once the Android app is completed, multiple robots will be controlled with tablets that communicate via a wireless base station at the pool.

“Unfortunately, Switzerland is not blessed with ocean access and thus does not have a strong industry interested in ROVs [remotely-operated vehicles, such as the Giant Crabster] or AUVs [autonomous underwater vehicles, such as Grace],” says Siegenthaler. “We are searching for possible supporters and investors. But for now, we build the naro-nanins for schools and we pursue research projects with naro-tartaruga.”

Students are encouraged to design their own robot fish with the naro-nanin edutainment pro...

If there’s demand, the project could be commercialized. The naro-nanin robot could cost as little as US$2,000 to $2,500 depending on what is included in the final specs, and its production costs. Naro’s other projects, like the naro-tartaruga, are working towards autonomy and may help the local police to detect objects at the bottom of lakes.