Finally, AI Can Smell
Olfactory Developments in Artificial Intelligence
Artificial Intelligence has mastered a wide spectrum of abilities by now.
AI is everywhere around us today in form of face recognition, voice assistance, maps, predictions, etc. However, there was one realm where AI lacked the right skills till recently. Now it’s getting better at it and the practical uses are numerous.
Making machines recognize smells was a concept that failed multiple attempts
The reasons for the failures were many. The accuracy was low and the limited performance didn’t pave the path to any practical uses. The initial inventions involved high costs of development and complicated installations. Few of them which worked okay enough ended up being confined to industrial use and never reached the masses.
But the research still continued at labs. The ability to smell had played a crucial role in human evolution. It was believed that if this ability is mastered by AI, it would turn out quite helpful for mankind.
A company called Aryballe has recently come up with some interesting developments that are turning out to be major milestones in the world of odor analytics. They have introduced a product that can smell with great accuracy.
“Aryballe combines biochemical sensors, advanced optics, and machine learning in a single objective solution to collect, display and analyze odor data so companies can make better decisions.” — The Official Company website.
Like most other kinds of AI, Aryballe’s digital olfaction’s implementation mimics humans.
Animals recognize smells by first collecting the fumes through noses and then signaling the brain. Similarly, the first step in the digital olfaction process is the collection of the smell. The biosensors attached to the olfaction device perform this function.
After the smell capture, the backend software draws out an optical illustration to represent the smell. This drawing follows preset rules such that similar smells will have similar illustrations. Already recorded smells have their illustrations stored in the backend database.
The last step in digital olfaction is pattern recognition. The optical illustration of the new smell is compared to the existing illustrations stored in the database. This process eventually leads to recognition. As the system learns more odors, it masters the recognition process.
Aryballe is just one of the major players researching digital olfaction.
Machines can smell, so what?
Well, this ability is opening doors to many new practical use cases.
One major area that is being focused on is the detection of hazards in automobiles based on detecting odors of leaked fuels or gases. Just like how humans can recognize the smell of petrol, if the system itself can smell leakages and sent our early warnings, it can be a major step towards enhancing safety.
The ability to distinguish rotten food from good food helps all animals in their survival. Now developments are being done to see if this ability can be used in the food industry for the mass acceptance and rejection of edible products.
Another major development of digital olfaction is in the health industry for early detection of diseases by analyzing human breath. While not as useful as the above four, research is also going on to re-create ‘historic smells’.
Praiseworthy milestones in olfactory developments of AI have happened in the last two years. While it is too early to predict how fast these developments would hit the markets, the future is looking very promising.