We’re starting to see robots gain a foothold in the food industry in some pretty interesting ways, from droids that make deliveries, to systems that make 300 pizzas an hour to chefs. Manual controls operate the frying stations. Researchers at the University of Cambridge have tinkered with aspects of this robotics field and developed a machine capable of “taste-testing” food as it works, ensuring balance taste just the way it needs to be.
The robotic chef developed by scientists is essentially a continuation of a project we looked at in 2020, in which the Cambridge University team collaborated with local appliance company Beko to come up with an interesting concept. The idea is not only to have a machine that prepares pizza or burgers like we’ve seen before, but to create the best possible meal based on human feedback.
It’s clear that everyone’s taste is different, and to cater to the inherent subjectivity in crafting a good meal, researchers have developed a new type of machine learning algorithm. Providing robotic feedback from human samplers has helped it improve its product over time, tweaking methods, and whipping up a final omelette that “tastes great.”
Now looking to give the robot self-taste testing, the scientists have teamed up with Beko again to produce a new and improved version. In doing so, the team sought to mimic the process of chewing in humans, which not only helps break down food for easier digestion, but also floods our mouths with saliva and enzymes that make food easier to digest. change its taste.
Evolving over millions of years, this process also sees saliva carry chemical compounds from food to taste receptors on the tongue, sending signals to the brain to determine if something is present. delicious or not. If a robotic system could do something similar, it could quickly adjust its cooking process, eventually succeeding with a tastier dish with less human intervention.
“When we taste, chewing also provides continuous feedback to our brain,” said study co-author Dr Arsen Abdulali. “Current electronic testing methods only take a single snapshot from an already homogenized sample, so we wanted to simulate chewing and tasting more realistically in a robotic system to creating a better end product.”
The team’s new machine uses a conductivity probe as a salinity sensor, which is fixed to a robotic arm. The robot was then presented with nine different variations of scrambled eggs and tomatoes, with varying amounts of tomato and salt in each dish.
The robot was able to “taste” the meal, with the dishes then passed through the blender multiple times to mimic chewing and allowing the robot to continue to taste flavors at different stages of the process. The various readings taken by the robot allow it to create a grid-style flavor map of dishes, based on how salty different “bites” are.
The scientists hope to add even more functions to their robot chef, planning to study new sensory abilities that will allow it to taste sweet and oily foods.
“When a robot learns to cook, like any other chef, it needs cues to show how well it did,” says Abdulali. “We want the robots to understand the concept of taste, which will help them cook better. In our testing, the robot was able to ‘see’ the difference in food as it was chewed, which improved its ability to taste. “
The study was published in the journal Borders in Robotics and AI.
Source: Cambridge University via EurekAlert