According to the Environmental Protection Agency, a majority of rivers and streams in America cannot support healthy life with the number of rivers being polluted trending upwards.  55 percent of waterways in America are currently listed as “poor” and another 23 percent are listed as “fair.”  Additionally, millions of Americans drink water that contains unsafe levels of industrial chemicals according to Environmental Science & Technology Letters.  When it comes to water pollution specifically, AI technology can help detect the sources of pollution for clean-up.

This week’s AI spotlight is on a robot called Envirobot which uses AI technology to find sources of pollution in bodies of water.  Developed by scientists at the Swiss research institute, École Polytechnique Fédérale de Lausanne (EPFL), Envirobot is a four-foot long eel-like robot that is made up with small compartments attached to each other.  Each compartment serves a different purpose – electrical, chemical, and biological.  The head of the robot houses the camera and computer.  Envirobot slithers around bodies of water to predict and find the source of pollution.  The compartments enable the robot to conduct various tests in the water.  It can both move to follow a waypoint placed by humans or move autonomously through the water on its own mission.   Then, Envirobot collects and relays the data that it finds.

While Envirobot does not seem like a very sophisticated use of artificial intelligence/automation technology, the applications for Envirobot have serious potential to help with water clean-up and pollution prevention.  According to EPFL’s Auke Ijspeert, the head of the lab that created Envirobot, swimming robots like Envirobot work more efficiently than conventional underwater robots which are more likely to get stuck in debris.  Additionally, swimming robots disperse less pollutants than traditional robots when moving around.  In the future, the lab hopes that they’d be able to utilize more than one robot in bodies of water and soon thereafter get a detailed map of the water of different biological and chemical factors.

Automation and AI technology is helping solve many environmental problems around the world.   Many companies are looking into how to use AI to predict and treat pollution all over the world.  In Envirobot’s case, it’s important to note that Envirobot strongly relies on human interaction in order to function.  Humans are necessary to deploy it and collect and interpret its data.  Uses for robots like Envirobot also present new opportunities in robotics to develop these kinds of devices.  This is in line with SIIA’s view that AI technology and automation largely add new job opportunities in areas where we may not even be able to perceive yet.

Considering the wide-spread pollution of water ways in the United States, people have been searching for more answers on how to solve these problems, perhaps even before they begin.  Imagine an Envirobot searching around a water way and discovering pollutants to be treated preventing a city using that water-way for drinking water from becoming the next Flint, Michigan.  This is not just an environmental issue, it’s a health issue.  Perhaps even in the future, these types of AI-utilizing robots would be able to treat the water themselves.

Diane Pinto is the Public Policy Coordinator at SIIA. Follow the Policy team on Twitter @SIIAPolicy.