According to a recentTechXplore, robots are making huge strides in a variety of industries, including some quite unusual ones. Currently, researchers at the Idiap Research Institute in Switzerland, the Chinese University of Hong Kong (CUHK) and the Wuhan University (WHU) have developed a machine learning-based program to instruct robots to stir-fry like professional chefs.
Intelligent robots that can prepare food
"Our recent work is the combined collaboration of three labs, the robot learning and interaction group led by Dr. Sylvain Calinon at the Idiap Research Institute and the collaborative and Versatile Robots laboratory led by Prof. Fei Chen Cuhk," said one of the researchers who conducted the study. Junjia Liu, a member of TechXplore.
"Our three labs have been working together for over ten years. We have an interest in developing intelligent robots capable of preparing meals for those who don''t."
The first research aims to establish a robotic chef, something that has been very difficult to achieve.
"While domestic service robots have been significantly developed in recent years, creating a robot chef in the semi-structured kitchen environment remains a huge challenge," Liu said.
"Food preparation and cooking are two key activities in the house, and a robot chef that can follow arbitrary recipes and cook automatically would be beneficial and a unique and engaging experience."
To complete a difficult task like stir-frying, Liu and his team first had to develop a bimanual-koordinator known as a "structured-transformer." This experiment was followed by human demonstrations.
"This mechanism treats coordination as a sequence transduction obstacle between the arms'' movements and adopts a combined technique of transformer and GNN to achieve this," Liu said.
"Thus, in the online process, the left-arm movement is adjusting according to visual feedback, and the corresponding right-arm movement is generated by the pre-trained structured-transformer model based on the left-arm movement."
Cooking both at home and in public
Liu now hopes that his new and improved model will one day transform into robots capable of making meals at home and abroad. It may also be used in developing robots capable of performing other tasks such as using two arms and hands. This already-popularpizza-making robot is one excellent example.
"We will now use higher dimensional information to develop more humanoid motion in kitchen skills, such as visual and electromyography signals," Liu said.
"We only used the relative displacement as the desired target," says the author. We also intend to propose a more comprehensive framework that consists of both the movement of bimanual manipulators and the state change of the object."
The results of the study were published in the journalIEEE Robotics and Automation Letters.
This paper demonstrates a method to develop well-known Chinese cooking art stir-fry on a bimanual robot system. This sequence of highly dynamic coordinated movements is often difficult to learn for a chef, yet it is typically difficult to dissect it. Finally, we propose a decoupled framework for learning this deformable object manipulation from human demonstration. First, the dualarms of the robot are decoupled into different roles (a leader and follower) and each one is evaluated separately, thus the bimanual