Cookies in the oven, freshly cut grass, the ocean breeze. What’s your favorite scent? Personally, I’m a huge fan of those in candle form. My top three at the moment have to be sandalwood, verbena and lemongrass, and lavender. With candles, larger ones emit stronger scents that can fill a room in minutes. But how does the scent move in space? A team of scientists is trying to find out.

Odors are typically invisible. However, a study being conducted at the University of Colorado Boulder (CU-Boulder) is attempting to visualize these odors. The National Science Foundation (NSF) has awarded a $6.4 million grant for “olfactory navigation research” to CU-Boulder professor and fluid mechanist John Crimaldi and his team. The grant was awarded as part of the BRAIN (Brain Research through Advancing Innovative Neurotechnologies) Initiative. Crimaldi and his team—including contributing scientists from six other universities—constructed a 50-foot-long, 5,000-gallon water tank in CU-Boulder’s Engineering Center for this experiment. After turning off the lights of the Engineering Center at CU-Boulder, a network of high-powered lasers are used to light up the bottom of the tank. Then, a dye is added as “surrogate” for an odor. When the lasers react with the dye, what looks like green flames emerge—these flames show how a scent would move.

This and other related experiments are the basis of the odor navigation project. The purpose? To teach robots to smell. It may seem far-fetched, but Crimaldi points out why it shouldn’t. Technology can already emulate other human senses. (For example, cameras can mirror sight with facial recognition software, while speech-to-text software and cochlear implants parallel hearing.) So why not smell? Smellbots, as they’re being called, could be useful for locating the source of an odor. Currently, humans often rely on canines for this—to locate contraband, explosives or people. This can put the dogs in danger and can be inconsistent depending on the dog and what it is tracking.

The first step to create a smellbot is to understand how odors move in space. The visual patterns of smell created at CU-Boulder form the basis for mathematical formulas explaining how scents move. The team is also focusing on how animals extract location clues from odors and how they use movement to enhance these odor clues while advancing toward their target.

After the completion of this study we’ll hopefully have a better understanding of how smell works. With these results, robots with the ability to smell may be not that far in the future.

Did You Know?

Bloodhounds, renowned as the dog breed with the greatest sense of smell, can distinguish scents over a thousand times better than humans. A bloodhound’s nose contains around 230 million scent receptors—40 times the number in humans!