Reach Robotics is closing
techcrunch.com
techcrunch.com
Anki recently learned the same lesson. People might pay $300 for a game console but they won't pay it for a single game. Even if your toy robot has half a dozen modes, it still only plays one game: 'toy robot'.
Mekamon looked like it had some impressively agile movements, so there was actual robotics there, but I don't think it did anything with them that you couldn't have done with an R/C truck or tank.
What I would really pay a lot for: Small hunter-killer robots to hunt insects in my home!
A bunch of small robots equipped with night-vision and some tiny weapon(s), mechanical and/or laser, crawling around catching bugs. Sure, spiders do that too - but I really, really don't want to live with (actual) spiders. It's psychological. A few well-armed mechanical spiders for bug hunting would be nice. I had bed bugs once, if I had had small insect hunter-killer robots well-equipped with sensors to see them crawl to me at night it would have been a godsend. Anyone producing something like that that actually works, you'll make billions! And it really doesn't have to be all that intelligent: Think about it, mosquitoes, bed bugs, they really don't have much of a brain, and how well they work. It should be doable to build robots that can compete with that low level of intelligence.
I suspect future attitudes to manual vacuum cleaning will be similar to manual clothes or dish washing.
Obviously even the salt pellet solution would be problematic if automated (say pest lands or is near your face or eyes).
However, a robot which could attack a pest niche where pests are endemic could have a chance at success.
Summer mosquitoes, Florida roaches, ants, stink bugs, etc.
- Weed the garden. (Done, or at least in progress)
- Keep mice out of my garage (Not done, seems like an obvious application)
- Clean my gutters (Likewise -- this alone is worth hundreds of dollars a year)
- Keep my driveway free of leaves and debris (does anybody make a "Broomba"?)
These are all applications that I'd pay for, and/or am already paying humans to do. I don't think generalized pest control via robotics is going to be workable, but certain specific pest problems like rodents seem like they would be easy to tackle.
But they can’t clean toilets.
Of the three companies listed in this article two of them were (expensive) toy companies, and the third was a crappy Alexa clone.
A Roomba solves a problem. Maybe not as well as we’d hope, but most of the time, it does a good enough job. Dishes and clothes can be cleaned mostly automatically. The other household tasks are hard to solve. I’m thinking dusting, taking the trash out, ferrying the clothes to and from the washer. Loading and unloading the dishwasher. I’d love to have something do these for me, but they’re very hard problems, and I don’t expect them to be solved in my lifetime. Until C-3PO becomes a consumer product, I don’t see this space making any money.
Restaurants and hotels do dozens of loads of dishes and laundry every day, so that additional expense and complexity required to automated the final 10% of these tasks is much easier to justify. So if you want to build a machine the folds laundry, don't build something ridiculous like the Foldimate, build a machine that folds bedding for a hotel.
But yes, hotels, prisons, hospitals, and similar places would be the initial target-markets for anything laundry-related.
Anyway, email me if you’d be interested in building this :P
Laundry is the domestic equivalent of de-mining: the perfect robot task if it weren't for the technical and financial challenges. You have a laundry robot at home already and it does everything except folding for a reason.
Here's a video of the state of the art not too long ago, using a $500K robot. Video is 50X speed.
The difficulty with home robotics is price vs capability. Capability is still much less than what is expect5ed via TV and movies. I think I can make a robot that sets the table up for breakfast/lunch/dinner, but just the hardware would cost the equivalent of a family car. A bit cheaper if I specialize the HW to its task only. But then it does only the table thing. Other tasks? More software, more general hardware. Sure, I can have a robot to do floor sweeping and another to make the table, but then I end up with multiple robots which share a lot of similar HW.
The lowest hanging fruit was dishwashers, washing machines, and automatic coffee machines. And that fruit hung low because it still offloaded all the complex spatial topological perceptual dexterous stuff to the humans.
And robotics is 10X as hard.
Then make it a consumer product that no one needs. And snap, you're dead.
I think either it needs to be an amazing teaching tool and/or have other useful capabilities (like merging a roomba/educational robot, or merging educational/house monitoring); or the cost needs to come way down for mass adoption. I believe both have future (my dream is widespread robotics within <$10).
Getting it to come up the stairs shouldn't bee too hard, right?
The problems with consumer robotics rarely are to do with the actual mechanics. Most electrical undergrad degrees have a module where you program a robot to follow a line or course, and it's quite simple to do. The problems occur when you take it out of it's test environment and 'leave it to the elements' as it were.
I'd say when that is achieved, the killer feature would be privacy.