Washington University thinks bugs could help find bombs

ST. LOUIS -- In the basement of Washington University's Brauer Hall, there's a room full of locusts -- hundreds in each dimly lit cage behind a glass panel, separated by age.

The colony, researchers call it.

"I like to joke that I could release the plague on St. Louis," said Nalin Katta, a doctoral student working with the bugs.

Locusts' antennae act as a nose, with sensors more complex than any chemical sensor an engineer could make.

A group of researchers at Washington University hopes to hijack that sense of smell, using bioengineering to create the ultimate smelling machine.

By using sensors that monitor locusts' brain activity, researchers would be able to tell what chemicals the locust is smelling.

The possibilities? Picture a swarm of locusts sent out on a mission to sniff out bombs.

Researchers picture the bugs helping the military detect land mines or verifying bomb threats at airports.

If all goes as planned, bionic bugs could join the ranks of police dogs and land-mine sniffing rats in detecting explosive devices.

"The biology of this beats engineering hands down," said Baranidharan Raman, head researcher on the project.

"So, why not, instead of trying to make a new device, we use the biology?"

Raman, associate professor of biomedical engineering, has spent his career studying locusts' sense of smell.

He just received a three-year, $750,000 grant from the Office of Naval Research to further his studies, perfecting these biorobotic smelling machines.

Using the grant, Raman hopes to combine three ideas.

First, researchers will implant sensors in locust brains to understand neural activity when the bugs smell different things.

They'll use an algorithm to interpret brain patterns, allowing them to decode what the locusts smell.

In past research, Raman has found locusts can be trained quickly to recognize different scents.

Taking a page from Pavlov, Raman and his team hit locusts with a puff of a smell, then reward them with a grass pellet.

Within five or six trials, a locust learns to associate that specific smell with food.

Second is creating a way for that information to go from a flying locust to its human counterpart.

That's where Shantanu Chakrabartty, professor of computer science and engineering, comes in. He is creating a tiny "backpack" for the locust to wear that would transmit the bug's brain activity back to researchers.

But it can't be too heavy.

Sturdy as they are, locusts can't handle the weight of a battery.

So, Chakrabartty plans to use the energy of the locusts' own movement to power the backpacks.

The backpack is in the prototype stage.

Last comes being able to steer the locusts.

To do this, Srikanth Singamaneni, associate professor of materials science, tattoos the locusts' wings with a biocompatible silk that conducts heat.

Apply heat to the right, and a locust veers left.

Apply heat to the left, and the locust veers right.

Steering the locusts has been tested successfully, Raman said.

Researchers have yet to figure out whether it would be more effective to steer the locusts toward a potential threat or let them sniff it out themselves, Singamaneni said.

For now, the focus is on the technology.

Logistics will come with testing.

The silk tattoo also can absorb chemicals in the air, allowing scientists to verify what the locusts smell.

The three aspects of the project all have been demonstrated individually. Now, it's just a matter of combining them all.

Raman hopes to be running complete lab tests within a year.