Robotic furniture movers

As Richard’s comment from yesterday confirms, there would be a real use for furniture that can rearrange itself.

Alas, the economics just don’t add up. The cost of equipping each table and chair with the appropriate motors, electric power, control electronics and wireless communication would be prohibitive. Decent office furniture is already too expensive. Robotic furniture would be far more expensive.

But there is another way.

Rearranging your furniture overnight does not require real-time performance. It just requires that the job get done over the course of hours — say, between the hours of 2am and 6am. Which means that your various pieces of furniture don’t need to move at once. They just need to end up in their proper places by the end of the night.

If the tables and chairs are built with a few design constraints — such as being easy to roll and having legs that can be readily gripped by a low-to-the-ground robot (say, four to six inches off the ground) — then all the work could be done by a single robotic furniture mover.

This would, in a sense, be a cousin to the Roomba, but rather than pick up dust, it would move around tables and chairs.

Given the advanced state of today’s computer vision and path planning algorithms, I don’t really see any engineering impediments. During the day the robot would park itself in an out-of-the-way corner and recharge. After everyone is gone for the night, it would get to work.

Going on-line to schedule a conference room or classroom would be pretty much the same as it is now, except that you could also specify the kind of furniture arrangement you’d like for that day’s workshop, lecture or discussion group.

What could be easier? 🙂

One Response to “Robotic furniture movers”

  1. PhilH says:


    To automatify your furniture, just get some 2d barcodes and slap them on the attach points; table legs, anchors, etc. All you need are some Kensington lock style anchor points, and a generic bot can shuffle things around to your heart’s content.

    I think the path we have to assume for all this kind of change is an incremental one; how do robots move our existing furniture, or mild variants thereof.

    Once robotics is small enough and cheap enough, I think there will be a small set of ubiquitous form factors; say Skutters, Quadcopters, Autocars. Everything else will be adjusted for integration with them. We saw it in computing; so many architectures are possible, yet the world is filled with x86+ and ARM. The network effect is there.

    And the consumer bots will outnumber the specialised industrial bots 10,000 to 1 because ubiquity implies value for money. Today, even supercomputer clusters are built with the same architecture that sits on your desk, because it’s well-developed and cheap.

    Your next house may well have a hatch for size Q43 quadcopters.

Leave a Reply