Sewer Robots Sift Data From Raw Human Waste

Meet Mario and Luigi—two bots on a mission to improve public health, one excrement sample at a time

MIT Senseable City Lab
Luigi (right) is the successor to Mario (left), the first MIT sewage-monitoring robot
Advertisement

I’ve always had a thing for Mario’s brother Luigi. Maybe it’s his height, or his ghost-sucking vacuum cleaner. So when I recently heard of another Luigi—also tall, thin, and willing to drop into sewers—I had to meet him.

At MIT’s Senseable City Lab, a team of researchers recently premiered a second-generation robot named Luigi who sifts through sewage as a way to improve public health. Luckily, I didn’t have to brave the smells of the sewer to meet him.

In a small campus makerspace stacked to the ceiling with batteries, wires, and glue, architects Newsha Ghaeli and Alaa AlRadwan introduce me to Luigi, then proceed to do what all good inventors do: repeatedly try to turn the robot on.

A long, tube-shaped device about the length of my arm, Luigi is the successor of Mario, a larger, chunkier robot that was the first to plunge into the sewers of Cambridge, Massachusetts. The robots are part of MIT’s Underworlds project, which we described last year as an initiative to create a “smart sewer” able to track microorganisms and chemicals that pass through the human gut and into the sewer system below, including bacteria, viruses, and drugs.

“We can really tell a lot about a person by sampling their gut, and all of this data is getting flushed down the toilet,” says Ghaeli. In their second major experiment to try to access that data, ten Luigis will be carefully dangled beneath manhole covers around Boston and Cambridge this summer in an effort to explore the diversity of sewage in different neighborhoods.

Begun in 2014, the Underworlds project is the brainchild of two MIT professors: urban architect and engineer Carlo Ratti and microbiome expert Eric Alm. Together, they believe real-time data collected from sewage could allow health officials to, say, detect an outbreak of flu or a spreading foodborne illness, or know when a population is stressed out or becoming obese.

It’s not the first research project to tap into sewage for public health information, but most such efforts rely on manual samples from sewage treatment plants, which are typically miles away from their source. Boston’s sewage, for example, takes two to three days to flow out to a local sewage plant. During that time, populations of microbes or concentrations of chemicals in the sewage can change dramatically. The Underworlds robots, on the other hand, are close to the source, and will hopefully soon transmit data in real-time, providing an instant snapshot of the health of a community.

The robots have come far in less than a year. “We first started with a big pole with a bottle duct-taped onto the end of it,” says Ghaeli with a laugh. Their first real effort was Mario, a heavy, hexagonal robot with six compartments to suck up sewage samples via syringe. But Mario was difficult to move around; the pressure of the syringes appeared to make bacterial cells explode; and it was pain-in-the-ass to clean. And there was that time the battery packs caught fire. So they created Luigi.

During my visit in the lab, AlRadwan eventually finds the correct remote to turn the bot on, and Luigi begins to hum as it lowers itself toward a bottle of liquid, our “sewage” for the day. Thanks to an ultrasonic sensor, the cylindrical robot stops 40 centimeters above its target. Then a quiet pump kicks in, and the bot begins sucking up liquid: first through a filter designed to stop toilet paper and other unpleasant sewage clumps from entering, then through a tube and into a camping water filter that processes the liquid, trapping viruses and bacteria inside. That camp filter can later be popped out of the device and taken to the biology lab for analysis.

In the future, the team hopes Luigi will run autonomously (without a human-operated remote nearby) and include sensors to detect pathogens on the spot, without needing to send samples back to the lab.

Perhaps Luigi’s greatest asset is its ability to self-decontaminate. Mario required a lengthy decontamination process, including exposure to ultraviolet light. But Luigi simply pumps a bleach solution through its system. A quick bleach spray on the plastic exterior, and the bot is ready to get back to work.

In addition to the 10-location experiment planned in Boston this summer, the team will soon publish the results of their work last year monitoring how sewage patterns in Cambridge change during the day, such as when one can detect more toilet water than shower water.

And in May, the team is taking their work to Kuwait, where Ghaeli, AlRadwan and other team members will invite colleagues to join them at a sewer for a demonstration of how Luigi could help their city monitor urban health patterns. “The first obvious app is an early warning system for viral outbreaks,” says Ghaeli. “But we’re designing this to be as flexible as possible, so we can put any sort of analytics tools on it.”

The Human OS Newsletter

Biomedical engineering in a biweekly newsletter. Expert insights into wearable sensors, big data analytics, and implanted devices for personalized medicine.

About the Human OS blog

IEEE Spectrum’s biomedical engineering blog, featuring the wearable sensors, big data analytics, and implanted devices that enable new ventures in personalized medicine.