A deep-sea underwater imaging and sensing system made using a Raspberry Pi Zero and Camera Module is making a splash in marine exploration circles. In the latest issue of The MagPi, David Crookes dives in.

It would seem there are few limits to where Raspberry Pi can go. The computer has been used in the Antarctic and on the International Space Station. It has also, as this project proves, plunged deep underwater – in this case, furthering scientific knowledge and enabling more people to explore the world’s oceans.

Indeed, without Maka Niu – a small, low-cost, deep-sea image and data collection system that has Raspberry Pi at its heart – Katy Croff Bell, president of the Ocean Discovery League, says such exploration could continue to be the exclusive domain of the wealthy. “By making this technology open and available to anyone,” she says, “we hope to bring more equity to this important field.”

This has far-reaching benefits. Less than one percent of the deep sea-floor has been seen and characterised, while most coastal countries lack the tools needed. “Ultimately, more observations will result in more discoveries and a better understanding about the single largest biosphere on Earth,” she adds. 

Not bad for a project that was initially earmarked for the Festival of the Pacific Arts and Culture, aimed at teaching high school students about deep-sea exploration and research.

Sensing the ocean

“The idea for Maka Niu had been brewing for a few years while I led the Open Ocean Initiative at the MIT Media Lab,” Katy explains. “The spark was a trip to Hawaii in January 2020 where my team and the MIT Future Ocean Lab were working with the Polynesian Voyaging Society and decided to create a small, low-cost camera.”

The festival was cancelled due to Covid-19, but Katy says this allowed the group to take a step back and work on making the end product even better than originally envisioned. “In the last five to ten years, there has been a growing movement towards the development of lower-cost systems, but still many of them can’t go very deep (tens or hundreds of metres) or they can only collect one type of data such as imagery,” she says.

“Maka Niu can collect several types of data including video and still imagery, depth, temperature, and GPS location when out of the water. It can go up to 1500 metres below sea level, and we have designs to go 6000 metres which would enable it to reach 99 percent of the sea floor.

“It’s designed to be modular so you can create other types of sensors or modules using the same housing, battery pack, and electronics. And we’ve designed the interface so that it’s really easy to program, use, and see your video stills using a smartphone or tablet without internet access.”

A deep mission

The use of Raspberry Pi felt natural. “The need for an imaging system that would be at least HD led to the adoption of Raspberry Pi Zero and Raspberry Pi Camera Module V2,” affirms Ocean Discovery League imagineer Dan Novy. “Brennan Phillips, from the University of Rhode Island, had created an epoxy potted [Raspberry] Pi-based camera called DEEPi which showed promise, and the cost and physical dimension of Raspberry Pi Zero were perfect in keeping the device small but powerful.”

Dan says there’s a Node.js-based mission engine accessed over Wi-Fi from a browser on a phone, tablet, or laptop. “Each Maka Niu acts as its own access point for direct communication between the browser and the mission engine, or the device can join a Wi-Fi network in client mode and, again, become accessible via browser-to-program missions or to review and upload footage in the cloud or download it to local storage,” he adds.

“The simplified programming environment allows the unit to be programmed in a LEGO-like fashion and allows non-programmers to be able to visually create missions just by stacking blocks of certain tasks and triggers into a loop that runs perpetually.” 

In the future, the idea is to use AI to automatically view and annotate mission footage. “We are working with CVision AI, scientists at the Monterey Bay Aquarium Research Institute, and others, on developing machine learning algorithms for automated detection and classification of organisms and other things,” Katy explains. “The collaborative design is just one step along a long-term path of continued co-development.”

The MagPi #125 out NOW!

You can grab the brand-new issue right now from Tesco, Sainsbury’s, Asda, WHSmith, and other newsagents, including the Raspberry Pi Store in Cambridge. You can also get it via our app on Android or iOS. And there’s a free PDF you can download too.

You can also subscribe to the print version of The MagPi. Not only do we deliver it globally, but people who sign up to the twelve-month print subscription get a FREE Raspberry Pi Zero Pico W!