Using robotic swarms to detect miniature narcotics submarines Thursday, 16 November 2017

In the US, narcotics traffickers are using home-made fibreglass submarines to import massive amounts of cocaine, heroin and other illicit substances into the country. Engineers at Boston University are developing swarms of seafaring robots that can better detect underwater threats like these.

Since the early 1990s, at least 60 narco-subs have been captured by anti-drug authorities. These submarines are sophisticated, with multiple crew members and GPS navigation. While there may be hundreds of these submarines out there, only a few of them are caught due to the difficulty of detecting them.

Currently, US authorities detect narco-subs through towed sonar arrays. This method can only be used in relatively deep water and can detect sound only within a limited range. Another method is to use sophisticated underwater robots called Bluefins to listen for targets, but they can only detect targets within a few kilometres, and cannot easily communicate their discoveries to human operators while submerged.

Greg McDaniel, a Boston University College of Engineering associate professor of mechanical engineering, is using a grant from the US Naval Sea Systems Command to develop new seafaring robots that can patrol large areas together, and collaboratively hunt for these submarines.

These pint-sized surface-dwelling drones would travel around an area of open water independently, listening for suspicious sounds. If an individual drone hears a noise, it alerts its peers, and they can collectively hone in on the target. This approach, which effectively makes use of a swarm of collaborating sensors, will make it easier to find and track targets, especially against the varying background noises of the ocean.

The research by the team is also expected to help contribute to emerging autonomous vehicles, monitoring stations and surveillance networks that are being developed for undersea distributed network systems.

The prototypes of the drones are built from off-the-shelf parts, with hulls made from foam boogie boards, the electronics housed in a waterproof plastic box that is strapped onto the board. The drones are propelled by a pair of commercial sump pumps, which spray jets of water backward through nozzles borrowed from fire extinguishers.

To detect underwater noises, the team have a store-bought hydrophone dangling on a cable underneath each drone. The electronics consist of a tiny computer board which processes sounds and picks out suspicious patterns.

The relative simplicity of the drones’ hardware allows the researchers to focus on the real work, which is the code that controls the swarm, allows for navigation and sound detection, and empowers the drones to detect suspicious sounds and triangulate their source.

“We’re really exploring ideas coming from the control and signal processing community,” he says. “How do we deal with all the acoustic data coming in? How do we compress it and share it? And then, how do we move each boat independently to get better data that helps them map a region or find and follow a target?”

The team is currently working on software and algorithms that let the drones map sounds around them, like large ships or underwater machinery, which could interfere with their ability to hear the much softer whirr of mini subs. Once a drone figures out where these loud sounds are, the team’s algorithm lets it send that information instantly to other drones, so they too can screen out those sounds.

With the drones remaining at the surface, they are able to rapidly and wirelessly communicate with each other through standard Wi-Fi technology. This allows them to effectively form a mesh network that allow the robots to be spread out over a wide area of the ocean, while remaining in contact with every other member of the swarm.

[Image: Greg McDaniel (left) pulls an autonomous robotic drone out of the Charles River lagoon as Noah Bernays (ENG’18) records data on his laptop. Photo by Jackie Ricciardi.]