Sometimes, a press release comes across the newsfeed that makes us stop, stare, giggle, and decide we have to share it with all of you. This is one of those stories.

Researchers at the University of Cambridge have been working with appliance manufacturer Beko to train a robot chef to taste food. But it’s not just about tasting food. No, they’re training the robot to keep tasting the food as it mashes it up, simulating how the taste changes as you chew. The results were published in the journal Frontiers in Robotics & AI and are hoped to help aid in the development of automated food prep.

After all, you want your robot chef to be a decent chef, and since human chefs taste as they go, it made sense to do the same with a robot chef. However, first, you have to train the robot to know what is pleasing to the human palette. In this particular research, the team focused on the saltiness of the dish prepared – scrambled eggs and tomatoes. According to the press release, the robot “tasted nine different variations…of scrambled eggs and tomatoes at three different stages of the chewing process, and produced ‘taste maps’ of the different dishes.”

As for how the robot tasted the food, that involved a conductance probe attached to the robot’s arm. The probe then acts as a salinity sensor, which the robot moves in a grid pattern to take different “tastes” of the dish. Not surprisingly, this new method helped improve the robot’s ability to assess the saltiness of the dish against the previous method, which just tested a single sample one time.

Don’t go expecting to buy your Star Trek-style food replicator any time soon, though. These robot chefs still need to figure out other flavors like sweet and oily.

More Information

University of Cambridge press release

“Mastication-Enhanced Taste-Based Classification of Multi-Ingredient Dishes for Robotic Cooking,” Grzegorz Sochacki, Arsen Abdulali, and Fumiya Iida, 2022 May 4, Frontiers in Robotics and AI