Cambridge University researchers invented a robot that watches cooking videos and learns to cook based on them. The researchers programmed their robot chef with a cookbook of eight simple salad recipes. After watching the training video, the robot was able to identify which recipe was being prepared and prepare it.

With the help of videos, the robot gradually updated its cookbook. At the end of the experiment, the robot also came up with a new recipe. The result of the researchers’ experiment showed how video content can be a valuable and rich data source for automated food production and enable easier and cheaper deployment of robot chefs.

Cooking is a difficult problem for a robot. Several commercial companies have built prototype robot chefs, although none are currently commercially available and they lag far behind their human counterparts in terms of skill.

Human chefs can learn new recipes by observation, whether it’s watching another person cook or watching a YouTube video, but programming a robot to cook different dishes is expensive and time-consuming.

“We wanted to see if we could teach a robot chef to learn in the same incremental way that humans do – by identifying ingredients and how they go together in a dish,” said Grzegorz Sochacki from Cambridge’s Department of Engineering.

Source: Science Daily