Australian scientists are robot able to distinguish alcoholic beverages only by their smell, And no, Futurama’s popular party robot is not Bender: his name is I DO NOT KNOW (‘nose’ meaning nose when translated into Spanish) and is practically a nose that distinguishes between brands, origins and styles.

This Dr Steven Su, He leads the entire process of this creation, from the University of Technology in Sydney, Australia, specifically the School of Biomedical Engineering.

“Until now, detecting the differences between whiskeys required complex and time-consuming chemical analysis, either by a trained whiskey professional who could still be wrong, or by scientists in a lab.” water stated, Statement by Sydney University of Technology.

“Therefore, having a fast, easy-to-use, real-time whiskey evaluation to determine quality and uncover any adulteration or fraud can be extremely beneficial for high-end wholesalers and buyers alike,” he added.

This technology NOS.E also detecting illegal parts of animals sold on the black market, like a black rhino horn. It also has great potential for healthcare application and disease detection.

NOS.E system to distinguish between different types of alcoholic beverages

How does it work? Have eight gas sensors To detect odors, the Sensor array generates a single signal string based on the odor molecules in the sample.

As explained by the Daily Mail, this is a analytical computer (shown in the photo). Thanks to a machine learning algorithm, the key characteristics of different alcoholic beverages can be recognized.

Researchers led an experiment six different types of whiskey: three blended malts (barley roasted to make the drink) and three single malt whiskeys. With the red and black labels, they worked with Ardberg, Chivas Regal and Macallan’s 12-year-old.

Less four minutes, The University of Sydney robot was able to get the results of the drinks with an accuracy. As stated in the previous paragraphs, 100% for place of origin, 96.15% for brands and 92.31% for style.

The robot will navigate the oceans and help discover new marine species.

Looks more like a pill than a pill robot made with artificial intelligence (AI)however CARL-Bot, most device a research team from Caltech, ETH Zurich, and Harvard, promises to explore and explore the ocean with greater precision.

Unlike other robots and devices used by scientists to study the oceans, CARL-Bot offers a special feature: thanks to its artificial intelligence, its creators want it to be autonomous and able to move freely in the ocean region.

Not like a fish drifting in the sea, but monitoring water conditions and making use of reinforcement learning (RL) networks; Unlike traditional neural networks, they are not trained on a set of statistical data, but are based on experience accumulated over time.

The performance of the AI ​​was tested with simulations simulating vortices and They teach you how to take advantage of slow speed zones by evaluating information from your environment. This is because in the future CARL-Bot will rely primarily on data it collects using gyroscopes and accelerometers.

In addition to perfecting the artificial intelligence, the team began experimenting with the robot itself. After testing for the first time in a small, well-controlled environment, they’ve moved to a larger tank and plan to test the device somewhere in the future. Jet pool capable of producing horizontal flow, in which CARL-Bot can move vertically, or learn to be at certain depths with various currents.