An ‘Electronic Tongue’ Could Help Robots Taste Food Like Humans


AI systems are already equipped to process stimuli like touch, vision, aroma, and auditory cues. So, what’s preventing them from experiencing taste? A team of engineers at Penn State is working on a project that aims to develop an “electronic tastebud.”

This device, built with materials mere atoms in thickness, can identify various gas and chemical molecules. While it’s not ready to crave snacks at odd hours, the researchers believe that their innovation might eventually assist robots in formulating AI-driven diets, devising culinary menus, and even teaching humans to diversify their taste experiences.

Humans’ dietary choices aren’t just influenced by their nutritional needs. They’re heavily swayed by personal taste preferences. While these preferences help us steer clear of bad or potentially harmful foods, they’re also the culprits behind our inability to resist an additional doughnut or a piece of dessert.

This dynamic interplay between necessity and desire involves a depth of psychological understanding that robots don’t currently possess.

Saptarshi Das, a lead researcher and an associate professor specializing in engineering science and mechanics, shared insights on this innovative venture in a statement on October 4. Das is credited as a key contributor to the research, which was detailed in the journal Nature Communications, elucidating the creation of a robot that can “taste” molecules.

The team crafted their “electronic taste receptor” by merging chemitransistors – sensors made of graphene that can identify gas and chemical molecules – with molybdenum disulfide memtransistors, which emulate the functions of neurons.

These two elements collaborated effectively, leveraging their individual advantages to mimic the sensation of tasting molecular compounds.

For instance, when exposed to salt, this electronic tastebud recognized sodium ions, essentially “experiencing” the taste of sodium chloride. Remarkably, the technology is versatile enough to discern all primary taste categories: salty, acidic, bitter, sugary, and savory.

Theoretically, the researchers can configure similar arrays of graphene-based devices to replicate the diverse range of taste receptors – roughly 10,000 – present on the human tongue.