Skip to main content
  1. Home
  2. Emerging Tech
  3. Legacy Archives

MIT designs autonomous robot that assembles IKEA furniture

Add as a preferred source on Google

IkeaBotWe’ve already got robots for some of life’s most tedious tasks – there’s the Roomba for vacuuming, the Looj for cleaning the gutters, and Siri for writing our texts when we’re too drunk to use our thumbs. In the utopian future, we’ll have a robot for every stupid chore we hate doing ourselves, and thanks to the wizards at MIT, we can finally cross one more of these tasks off the list: assembling IKEA furniture. Thats’s right, MIT researchers have actually developed a robot that will assemble flat-pack furniture on it’s own, with minimal direction.

This revolutionary contraption was unveiled earlier this week at the IEEE International Conference on Robotics and Automation (ICRA) by a team from MIT’s Distributed Robotics Laboratory. But before you crap yourself from being so impressed, you should know that the IKEAbot isn’t just one robot – it’s actually a pair of them. Just like with humans, it takes two to assemble a Borgsjö TV stand. Going it alone will only lead to frustration, rage, and involuntary enrollment in an anger management program. So, in an effort to avoid any outbursts of hellish robotic fury, the team from MIT wisely built the IkeaBot from a pair of KUKA  YouBots, each of which have been retrofitted with special arms for screwing things in. Not only is this a wildly innovative idea, it also finally gives us an answer to the age old question of “how many robots does it take to screw in a table leg?”

Recommended Videos

Words don’t really do it justice, so check out the video below to get an idea of how it works. Skip to somewhere around the two minute mark to see it in action.

The hardware is pretty impressive, but in all honesty the software is where the magic happens. To begin, the only thing the bots are given is a set of CAD files that describe the furniture in terms of geometry, number of parts, screw holes locations. From there, they use a custom object-oriented symbolic planning system to figure out how the pieces fit together – without actually knowing what the finished furniture will look like. Aside from the CAD files, all the bots know is that every screw hole should be used and that no parts should be left over. Amazingly, even with such simple rules, the bots almost always end up with the right configuration . 

Unfortunately, the project is far from being commercially available, but let’s not kid ourselves here – if you’re buying furniture at IKEA, chances are you can’t afford a pair of 20,000-dollar autonomous robots to assemble it for you. For now, you’ll have to build stuff the old fashioned way: on your knees, with the instructions in the wastebasket. But if you’re patient, I’m sure we’ll have these badboys at our disposal in just a couple years.

Drew Prindle
Former Senior Editor, Features
Drew Prindle is an award-winning writer, editor, and storyteller who currently serves as Senior Features Editor for Digital…
Google wants Gemini to help build the next big scientific breakthrough
Gemini for Science pushes agentic AI deeper into real research workflows
gemini for science

Google is building Gemini deeper into the research workflow, starting with ideas, tests, and scientific literature.

At Google I/O 2026, the company announced Gemini for Science, an experimental suite built around agentic AI science. It targets the manual work behind discovery, including hypothesis building, computational testing, and literature review.

Read more
You can now walk through AI versions of real places with Google’s Project Genie
Text, Logo

Google is pushing its experimental AI world-building project into surprisingly realistic territory. The company announced that Project Genie can now use real-world imagery from Google Street View to generate interactive virtual environments, blending real locations with imaginative AI-generated styles.

At its core, Genie is what Google calls a “world model” — an AI system capable of creating explorable digital environments where AI agents, robots, or even users can interact naturally. Until now, those worlds were mostly synthetic. But with this new update, Genie can anchor itself to real places pulled directly from Street View imagery. This is actually where things start feeling like a glimpse into the future of simulation.

Read more
Google wants to reinvent your TV remote with Gemini and pointer controls
Computer Hardware, Electronics, Hardware

Google is making a bigger play for the living room, and this time, it is not just about what you watch — it is also about how you interact with your TV. At Google I/O 2026, the company revealed a fresh batch of updates for Google TV and Android TV developers, all centered around one idea: TVs are no longer passive screens sitting in the corner of your house. With more than 300 million monthly active devices across Google TV and Android TV, Google clearly sees the television as its next major AI battleground. And Gemini is now at the center of that strategy.

The company says Gemini is already helping users discover content through natural voice interactions. But Google now wants the experience to feel more dynamic and conversational, almost like searching the web — except on your couch. Instead of only surfacing static results, Gemini on Google TV can now respond with a combination of visuals, videos, and text snippets to answer queries. So if someone asks for a thriller with a strong female lead or a documentary about space exploration, Gemini pulls contextual recommendations directly from streaming apps and their metadata.

Read more