— Inventors and Remarkable People
Dyson working to perfect robotic vacuum cleaner
Sir James Dyson at the Sydney launch event for Dyson's new line of Airblade hand dryers
A launch event in Sydney, Australia, this week was the latest stop on Sir James Dyson’s world tour introducing his company’s new line of Airblade hand dryers. While the dryers and the new digital motor that powers them were the main focus, the billionaire British industrial designer also confirmed that his company is still working to perfect an autonomous vacuum cleaner.
There’s no hiding Dyson’s pride in his company’s new digital motor. And why not? Developed at a cost of US$40.9 million over a period of 10 years, the new 1,600-watt motor spins at an impressive 110,000 rpm. “No one else makes a motor that goes any faster than about 35,000 rpm,” Dyson says, pointing out that even a Formula 1 engine runs at only 19,000 rpm.
As well as the three new Airblade hand dryers, the mini motor can also be found in the company’s Digital Slim DC44 vacuum cleaner and will be showing up in other Dyson products in the future.
Could one such product be an autonomous vacuum cleaner? The reduced size of the motor, with its 85 mm (3.4 inch) diameter and improved efficiency would certainly make it an ideal candidate. But size, efficiency and suction are just some of numerous problems that need to be overcome – intelligent autonomous navigation being another major one.
Dyson says the company has been working on such a device for 15 years and Gizmag actually did a story on the imminent release of a robotic vacuum called the DC06 back in 2004. If the company had followed through with a launch it would have had bragging rights as the first to bring a robotic vacuum to market. But despite claiming the DC06 is more advanced than other robotic vacuums already available, the decision was made to hold off until the design was perfected.
“When we can do it properly, which of course we think we can, we’ll launch one,” says Dyson.
About the Author
Darren's love of technology started in primary school with a Nintendo Game & Watch Donkey Kong (still functioning) and a Commodore VIC 20 computer (not still functioning). In high school he upgraded to a 286 PC, and he's been following Moore's law ever since. This love of technology continued through a number of university courses and crappy jobs until 2008, when his interests found a home at Gizmag.
All articles by Darren Quick
I agree that intelligent autonomous navigation is the key. Most robotic vacuums cleaners just use a steering Algorithm that aren't prefect.
If you would have asked me how to solve this problem in 2004 I would have said to centralize the intelligence in something like a PC and feed the sensor data to it over WiFi. It would allow the system to create a 3d model of what your home looks like and send instructions to the robot based on it.
Now the intelligence to make these computations is possible on platforms that are closer to the cost and efficiency of Rasberri Pi. Mapping rooms could involve expensive sensors but it doesn't have to. it could use Algorithm like everything else to start and then just keep track of distance traveled before colliding with objects to create a simple map of household.
You could even speed up the process by preloading it with a floorplan and once that is complete it would eliminate the need for devices like virtual walls to contain it inside a single room. It would also trivialize problems like finding the charging dock on the other side of the house.
In theory you could designate a single room for it to clean using a phone or tablet and it would undock itself and drive there. You would only ever have to pick it up to empty it.
The technology to do this is mostly already cheaply available. Being able to navigate around a home is probably the most important challenge household robotics need to solve. Vacuum cleaners are only one application but there are numerous others.
I'm amazed more companies aren't trying to solve that problem especially now that mobile computing has made it technically feasible.
Just throw a small, focused kinect sensor in the front of the unit, and have it spin in a circle a few times in various parts of the room to make a 3D map of everything below a foot. Then it can use those maps to determine the most efficient routes from one place to another. This isn't rocket science, though I suppose it would take a visionary to actually make it happen.
$40mil motor and 15 years of research for vacuum cleaners? I see why their products are ridiculously priced. Very happy with my VR6170LVM Roboking which maps the ceiling and can actually go back 5 minutes later to do part of a room after going out a doorway. Pretty quiet too
I'm getting really tired of his proselyting. He took an industrial cyclone and made it domestic. That's not inventing, it's re-purposing. Then he offshored it to a cheap labour market in Malaysia. Yet he keeps taking the high moral ground.
When will he stop trying to present like a critical version of Steve Jobs? He got lucky, that's all.
Over 160,000 people receive our email newsletter
See the stories that matter in your inbox every morning