Researchers develop robot that can build an IKEA chair in under ten minutes

That's several hours faster than the average human

Scientists based in Singapore have developed a robot capable of putting together an IKEA chair in under ten minutes without the help of humans.

The robot can also read and comprehend the instructions, rather than having to be programmed.

Developed by researchers at Nanyang Technological University, the assembly robot uses a 3D camera and two robotic arms kitted out with grippers to pick up parts.

Professor Pham Quang Cuong and his team, who work at the university's school of engineering, used algorithms from three individual open-source libraries to power the robot and to assemble an IKEA chair.

In a controlled experiment, the robot was able to assemble the IKEA Stefan chair within 8 minutes and 55 seconds. Additionally, it only needed 11 minutes and 21 seconds to study the plans and locate the parts.

Yesterday, the researchers published the results of the study in an academic journal called Science Robotics. Assistant Professor Pham said the robot works with "precision" and is "more complex than it looks".

The scientist explained: "The job of assembly, which may come naturally to humans, has to be broken down into different steps, such as identifying where the different chair parts are, the force required to grip the parts, and making sure the robotic arms move without colliding into each other.

"Through considerable engineering effort, we developed algorithms that will enable the robot to take the necessary steps to assemble the chair on its own."

Professor Pham, research fellow Dr Francisco Suárez-Ruiz and alumnus Mr Zhou Xian said the robot could eventually be used in "industries where tasks are varied and do not merit specialised machines or assembly lines".

The system was able to assemble the chair after taking 3D photos of its parts and generating a map that showed where they were located on the floor. It then used algorithms and a motion pathway to perform tactile perception, grasping and execution.

Currently, the robot is already pretty advanced, but Pham revealed that the research is still ongoing and that it is likely to get more artificial intelligence features in the near future.

"We are looking to integrate more artificial intelligence into this approach to make the robot more autonomous so it can learn the different steps of assembling a chair through human demonstration or by reading the instruction manual, or even from an image of the assembled product," added the specialist.