The robot cook learned to taste and evaluate the taste of food during cooking 4.7

Researchers taught the robot cook to taste food during cooking, simulating the process of chewing. The method made it possible to accurately and quickly estimate the amount of salt in the dish, as well as to make “taste maps” that brought the robot’s sensors closer to human perception.

Scientists from the University of Cambridge (UK), in collaboration with the manufacturer of household appliances Beko, trained a robot chef to assess the salinity of a dish at different stages of cooking. Before that, the robot already knew how to cook an omelet based on the reviews of tasters. Now he was able to taste the food himself, imitating the process of chewing and making a “taste map”. The results of the study, published in the journal Frontiers in Robotics & AI, will be relevant for creating methods of automated or semi-automated cooking. In addition, the new approach significantly accelerated the assessment of the salinity of food compared to traditional methods.

Taste perception is a complex process to which the appearance, smell, texture and temperature of food contribute. Saliva formed during chewing helps to transfer chemical compounds to the taste receptors that send signals to our brain. At the same time, the taste changes as the food is chewed, which provides constant feedback to the brain.

Often we try a dish during the cooking process to evaluate its taste.

The existing methods of electronic tasting are based on the analysis of a single homogenized sample. Therefore, scientists sought to reproduce a more realistic process of taste perception in a robotic system.

To do this, the researchers attached a salinity sensor to the robot’s arm. As he prepared a dish of eggs and tomatoes, he “tasted” the food, getting readings in just a few seconds. To simulate the texture change that occurs when chewing, the scientists placed the pieces of the dish in a blender, and the robot re-evaluated the taste. The readings collected at various points of grinding food in a blender allowed him to make “taste maps” of each dish.

By imitating the human perception of taste, robots will be able to learn how to cook food that people like and adapt to their individual preferences. Understanding the concept of taste will make robots better cooks. Such devices will be in demand in nursing homes, boarding schools, hospitals and other organizations where human resources are not always enough to provide all people with a delicious and balanced diet.

In the future, the authors plan to equip the robot with other sensors that will allow it to assess the fat content of food, as well as distinguish between sweet and sour taste.


Learn More →