(Credit:Eric Paterson) Dogs are used by almost every local, national, and international security agency to track down everything from an ounce of pot to weapons to criminals and disaster victims. They have great sniffers and several breeds are particularly trainable for search and disaster work. But training dogs is expensive and time-consuming and the dogs' work often involves physical danger to them.
RealNose, a multi-institutional project sponsored by the Defense
Advanced Research Projects Agency (DARPA), hopes it will be able to
simulate dog-sniffing ability in an electronic sniffing machine. Biomimicry? This project is about as biomimetic as you can get. Eric Paterson, Senior Research Associate at Penn State's Applied Research Lab, says that to develop RealNose, the researchers will have to engineer olfactory cells from actual canine olfactory cells!
Some sniffing devices already been developed, but they work only for detecting a limited range of chemicals. Paterson has been studying the mechanisms and transport of canine olfaction with the object of coming up with an improved device. So far, he and a colleague, Brent Craven, have created a computational fluid dynamics model of a dog's airway, which enabled them to identify the single passageway responsible for the transport of odors to the olfactory region.
Paterson and Craven will work with other researchers at Penn, as well as several other university research teams, to identify the canine olfactory cells that respond to certain chemicals and to engineer their electronic counterparts.
Paterson says the the project is "high risk, high payoff." But if it succeeds and a marketable technology emerges, the RealNose "could be tailored for detecting drugs, explosives, chemical and biological weapons, and even certain types of cancer."