Need a personal chef? This robot could cook you a scrumptious meal for breakfast
News > World
[types field=’icon-for-post-information’ title=’%%TITLE%%’ alt=’%%ALT%%’ size=’custom’ width=’40’ height=’40’ resize=’proportional’][/types] [types field=’post-author’][/types] | Posted on June 15, 2020
Engineers at the University of Cambridge in the United Kingdom partnered with the domestic appliance company, Beko, and trained a robot to make an omelette. Using techniques learnt through machine learning, the robot can complete the entire process from cracking the egg to seasoning it and frying it. Results show that it has successfully and consistently cooked omelettes that “tastes great,” proving its culinary skills.
Robotics and artificial intelligence (AI) have progressively evolved over the past few decades. Also, pop-culture has depicted many different AI-based assistants, some of which engineers have worked towards developing in real life. Of the many cultural depictions of robots, a robotic chef has always been a universally desired invention.
An article published by Cambridge notes that robotics companies around the world have attempted to construct their own versions of robotic chefs, many of which may have gone unreleased due to shortcomings. However, this research is stated to target the concept of food quality optimization, so the robot can learn to tastefully replicate the work of its human counterparts. The results of the research can be found in the journal IEEE Robotics and Automation Letters.
In the article, Dr. Fumiya Iida, an engineer at Cambridge who led the research, highlights the problem roboticists have with cooking due to humans’ varying views regarding food and taste. As a result, it is particularly difficult for scientists to assess whether the robot has done a good job or not.
The article further explains that teaching a robot to prepare and cook food is difficult because it must deal with complex problems in robot manipulation, computer vision, sensing, and human-robot interaction, in order to produce a consistent end-product. It’s also noted that cooking is a qualitative task, while robots generally excel at quantitative tasks. Since taste is not universal, universal solutions don’t exist.
Dr. Iida goes on to talk about the fact that an omelette is a dish that is easy to make, but difficult to make well. As a result, the researchers considered it to be an ideal test for improving the robotic chef’s abilities and optimize it for taste, texture, smell, and appearance.
The experiment was conducted by Cambridge’s Department of Engineering in a test kitchen supplied by the partnered companies Beko and Symphony Group. The machine learning technique was developed by Dr. Iida and his team, and it applies the principle of Bayesian Inference. This principle, according to an explanation published by Brown University, is a statistical technique that states that prior beliefs should be updated as new data is acquired.
During the engineers’ experimenting and testing, the results showed improvements in the robot’s cooking, thanks to the machine learning algorithm, prompting Dr. Iida to say, when comparing the robot to human chefs, “the omelettes, in general, tasted great – much better than expected!”
Watch the video below to see a demonstration of the robot’s refined culinary skills.
[types field=’icon-for-post-information’ title=’%%TITLE%%’ alt=’%%ALT%%’ size=’custom’ width=’200′ height=’200′ resize=’proportional’][/types]