Personal computer -> Personal robot?

It would have to be able to negotiate stairs and curbs and get both itself and its load into and out of any car. Then there's the issue of what it does if it gets separated from whoever it is following.

That's the big question.

For $5,000 it would have to do a heck of a job. I can pay people to vacuum my house for a *lot* less.

Charging itself is fairly easy. At least in the US power outlets are easy to find visually. Putting a charging circuit into a robot is easy if the robot is big enough.

Why does it need a base station.

You're assuming a general purpose robot: a Rosie the Robot. The Roomba sells because it does one task. Eventually they will produce one that does it well.

Indoor navigation is fairly easy once you can use some vision processing linked to other sensors.

The main problem (as I see it) with indoor navigation is that the obsticles change. I wouldn't want my robot to run over my dog.

Also a crowded environment often puts limits on the robot's size. For example I have a hard limitation of a 16 inch width.

I've thought of a few different ways to deal with stairs with a large robot. I'm waiting on some tools before I try these out.

I don't even trust maids to dust my stuff. However, that would be a truely killer application.

Now you're talking about a smart house. I do believe it would be useful for a domestic robot to interface with such things.

On the other hand, picking up trash would require being able to differentiate between trash on the floor and a robot part on the floor (I'm sloppy).

If you like building, then build. If you can solve these problems, then worry about commercializing it.

-- D. Jay Newman

My wife is disabled and in a wheelchair. How does she and millions of others like her deal with stairs, curbs, and cars? Well, our home has no stairs and I don't think that a simple robot like I am talking about would need to negotiate stairs. Where there are stairs a lot of the time disabled people have ramps. Here in San Jose, CA our streets (at least most of the ones I've seen and my wife negotiates) have slopes at the intersection curbs so that disabled people can negotiate them. Most malls that are multi-leveled have elevators. Buses here in San Jose as well as the public transit system all have ramps that allow disabled access. People that have wheelchairs have cars and deal with the problems that incur because of that. We personally have a wheelchair ramp for our van. With all this said, I have built my wife a small remote controlled robot that she operates remotely in front of her. It has a simple basket on top and can carry around 35 pounds. I will get around to putting a follow-me mode into it one of these days (probably). I think that if the bot got separated, it could just stop and give a yell (either audible or visual) to its owner who would then be responsible to "hook back up". Oh, and I built it for a cost well under $500.

That sounds like a useful device but it doesn't sound like a "killer app". Everybody who is disabled is not in a wheelchair and many people who are in wheelchairs cannot get their chair into or out of a car without assistance. So the simple device you describe would be useful for that subset of disabled people who are in wheelchairs but are in sufficiently good health otherwise that they can load stuff into a car without difficulty, which I don't think is a large enough percentage of the population to reach any kind of critical mass.

I don't think it would be much help to the thalidomide victim I was talking to the other day who in a foot race could easily run me into the ground and I suspect could have kicked the crap out of me in a fight as well.

Now if it could carry box of books up a flight of stairs with two landings it would have been a godsend to another fellow I know who at 90 just isn't up to that anymore.

Many people quote something like this and compare the pretty much non-existant home robotics explosion to that of the PC explosion of the 80's / 90's and its continuing growth into homes.

But you have to ask yourself - are PCs really the driving factor of that? PC's have been popular with businesses since their inception and with the advent of usefull applications like spreadsheets back in the '80s. But in the home, there were just a smattering here and there, by no means were they ubiquitous like today (at least on the US). OK, some people bought PCs even though there was no apparent non-programmer/geek use for them, but the vast majority of homes did not have PCs.

What has really gotten PCs into the homes of late? It's the 'net, I think. So the killer app has really been e-mail and the web. Oh, and games. I'd venture that the vast majority of home computers are used almost exclusively for those purposes - communications, information access and shopping, and games. This year is supposedly one of the largest on-line shopping years ever and many retailers/e-tailers are seeing their e-sales outpase their brick and morter.

So you just need to ask yourself, what will do that for an equivalent mass movement to put robots in the home?

Some people will argue that it's already happened - automatic dishwashers, clothes washers and dryers, etc - normal home appliances. Clearly this is where Roomba is intending to make its place.

I don't know. And if I did, I doubt I'd say until my business model was firmly in place :-) But I suspect that like the 'net, it will sort've creep up on us over time while the enabling technologies are invented, become cheaper and cost effective, and allow truly useful machines to be created and purposed to become second nature in our every-day lives.

-Brian