Robot Puts Together Ikea Furniture

A robot in Singapore can successfully assemble a piece of IKEA furniture. But there's still a long way before it can match human abilities, let alone be a superior alternative.

The robot is the work of Nanyang Technological University students and is made up of two robotic arms and a 3D camera.

The furniture in question is the "Stefan chair." It's a very basic chair frame that in theory is extremely simple to construct, however, the main reason it is challenging to build is due to a few screws that need to burrow quite deep in to wood. It therefore takes a surprising number of forceful revolutions to get the screws to bite into the wood.

Planning Was Key to Finishing the Task

Constructing the chair is a one-person task. Contrary to some media reports, the two robotic arms approach is more about replicating a human's use of two arms and hands, rather than the marriage-challenging task of two humans working together.

The robot wasn't speedy by any means, taking 11 minutes and 21 seconds of planning, three seconds to gather together the parts, and then 8 minutes and 55 seconds for the assembly. That's certainly slower than most humans would take on the task.

The point is more about the process than the outcome. Assistant Professor Pham Quang Cuong, who oversaw the work, noted that "[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." (Source: edu.sg)

While the 11 minutes planning stage may sound slow, this was key to the task. The robot had to start by identifying the pieces laid flat on the floor, replicating a human having opened the flat-pack box. It then had to use algorithms created by the students to plan the most efficient way to use both arms to bring components into the correct position without the two arms colliding.

Assembly Line Could Be Next

The students will now explore adding artificial intelligence to the robot to see if it can figure out how to put the furniture together by "watching" a demonstration video or simply looking at photographs of the completed chair. (Source: reuters.com)

While flat-pack furniture is a useful test subject, the most likely practical use of the students work would be an assembly line robot that could be quickly reprogrammed to take on a task that doesn't justify the expense of a dedicated machine.

Below is a video of the robot in action.

What's Your Opinion?

Was this a sensible use of the students' time and resources? Are you impressed by the robot performance or is it pointless given it was slower than most humans? Should we worry about robots being able to perform tasks that previously seemed like they required human brainpower and agility?