Sniffer dogs, beware: Scientists give robots a sense of smell | Tech News

Thanks to groundbreaking research, robots have been given a sense of smell that will make sniffer dogs everywhere dread redundancy.

Researchers at Tel Aviv University in Israel have created a biosensor that allows machines to detect and identify odors.

The breakthrough is attributable to nature, as the team exploited the locust’s ability to capture and interpret scents through its antennae.

The desert locust’s antennae are connected to an electronic system that uses machine learning to detect and measure odors with a sensitivity normally only found in animals and insects.

“Man-made technologies still cannot compete with millions of years of evolution,” the researchers said.

“One area where we’re particularly behind in the animal world is the sense of smell.

“An example of this can be found at airports, where we have multi-million dollar magnetometers that can detect if we are carrying any metal equipment.

“But when they want to check a passenger for drug smuggling, they bring a dog to sniff him.”

read more:
Sniffer dogs detect COVID with incredible accuracy
How a nice boy with a big nose can help keep the lights on

Kossi, one of the sniffer dogs being trained to detect coronavirus from arriving passengers. Sample, found on September 22, 2020 at Helsinki Airport in Vantaa, Finland.  Lehtikuva/From Reuters Attention Editors - This image was provided by a third party. No third party sales. Not for use by REUTERS third party distributors. Finland is out. There are no commercial or editorial sales in Finland.
Sniffer dogs are even being used to detect coronavirus

How do sensors work?

It essentially aims to replicate how our sensory organs — such as noses and ears — receive different signals.

When this happens, they are converted into electrical signals that are decoded by the brain – allowing us to accurately identify different smells and noises.

This part of the process was the most challenging for the Tel Aviv team—connecting the biosensors—in this case, the locust antennae—to an electronic system that can decode the signals.

Professor Yossi Yovel from the university’s School of Zoology explained: “We connected the biosensor to smell different odors and simultaneously measured the electrical activity elicited by each odor.

“The system allowed us to detect each odor at the level of the insect’s main sensory organs.

“Then, in a second step, we used machine learning to create a ‘library’ of scents.”

Scents that the sensor can identify include lemon, marzipan, and various Scotch whiskies.

Sensors are added to the robot, giving it its own “biological nose”.

The hope is that such machines could one day be rolled out at airports and elsewhere to help identify explosives, drugs and diseases.

The findings were published in the journal Biosensors and Bioelectronics.

Source link