Slashdot Mirror
RFID-Equipped Robots Used as Guide Dogs
Roland Piquepaille writes "A professor in computer science at the Utah State University (USU) is building robots to help people with disabilities, according to the Utah Statesman in this article. The story, which is more focused on the professor than robotics, carries several anecdotes, such as an embarrassing voice recognition system. After a blind man cleared his throat, the robot misinterpreted the sound as a sign that the man wanted to go to the bathroom. Later, every time a man cleared his throat before speaking, the robot changed directions and insisted to guide him to the restrooms. Even if the article is entertaining, this project at USU is far more ambitious. In fact, they want to design RFID-enabled robots mounted on mobile carts which will welcome blind persons at the entrance of a supermarket and guide them through the store. I bet you'll never find those carts at a Wal-Mart store, but read more for other details, references and pictures about these RFID-equipped robots designed to help blind people."
Part of the benefit of having a guide dog is that it can be a true companion to a blind person, and many times more trustworhty than a robot. A dog isn't going to crash and require a reboot, and kibble is easier to figure out than rechargeable batteries.
That said, computers and robotics are a reflection of their creators. I guess I trust dogs a whole lot more. A dog can also provide protection against burglars. The benefits of dogs over robots goes on and on.
A professor in computer science at the Utah State University (USU) is building robots to help people with disabilities, according to the Utah Statesman in this article . The story, which is more focused on the professor than robotics, carries several anecdotes, such as an embarrassing voice recognition system. After a blind man cleared his throat, the robot misinterpreted the sound as a sign that the man wanted to go to the bathroom. Later, every time a man cleared his throat before speaking, the robot changed directions and insisted to guide him to the restrooms. Even if the article is entertaining, this project at USU is far more ambitious. In fact, they want to design RFID-enabled robots mounted on mobile carts which will welcome blind persons at the entrance of a supermarket and guide them through the store. I bet you'll never find those carts at a Wal-Mart store, but read more...
First, let's look in detail at the failure of the voice recognition system.
Of course, this is only a very small part of the project, which will deploy radio frequency identification (RFID) tags for use in robot-assisted indoor navigation for the visually impaired.
But they already built prototypes. And below are two pictures showing this RFID-equipped robotic guide (RG) for visually impaired people (Credit: Vladimir Kulyukin)
For more information, here is a link to Vladimir Kulyukin home page -- which is not always available. From there, you'll have access to various pages covering his research interests and his publications.
You might also want to read a paper named "RFID in Robot-Assisted Indoor Navigation for the Visually Impaired," available as a PDF document (6 pages, 124 KB). Here is the abstract.
The benefits of dogs over robots goes on and on.
I would have to disagree, a guide dog does not know where the library is, or the computer store, or your house, whereas a robotic dog will(does?) have a map and be able to direct you to the location. This robot sounds like a great idea, a good use of technology.
This sig is definitive. Reality is frequently inaccurate.
Guide dog. Not attack hound. ...Besides, what would happen when said soldier, hiding somewhere near the robot, cleared his throat?
all i can see is a cut and pasted original article, do the editors even read the submissions ?
i guess copyright infringment is only bad when its software right ?
I always read the last part of this guys submissions first. It scared me.
"designed to help blind people"
-
"Our store brand of coffee is cheaper, sir"
"Last time you had cheese on your list"
I can even come up with things lawmakers would think of, like forbidding Robby the Robot to suggest cigarettes or liquor.
The robots took out insurance policies on the blind men, and they're using every excuse they can to lead them to the bathroom where they can "accidentally fall" and be left unconcious in a running sink.
It's like the I-Robot movie, only worse! Well... it's like the moive, at least.
My Greatest Heist - Muisc partly inspired by the unbeatable Qwantz
The blind face two distinct but related problems: finding the destination, and getting there safely. A dog or cane is an excellent tool to get to the destination safely (in part because sighted people recognize them and give extra space), but they do nothing for the other problem.
I've helped several blind people find the location of something they were 10 feet from. They were pretty sure they were close, but didn't know where to go next. Their dog would keep them on the sidewalk, but had no idea that they wanted to enter the building not go past it.
I studied this issue (about 10 years ago), and at that time all technology to solve the second problem was much worse than a dog or a cane. However there was promise in technology to solve the first problem. In fact I said at the time (and other experts agreed) that there was unlikely to ever be technology that would replace a dog/cane. However technology was very likely to supplement those tools to help the blind get to their destination.
"The experiments illustrate that passive RFID tags deployed in the environment can act as reliable stimuli that trigger local navigation behaviors to achieve global navigation objectives." That is, you'll have to imbed an RFID tag into everything you want the robot to be able to see. This is a way around the very hard problem of a generalized object recognition system for the robot. It's a fine solution, if you're blind and you don't mind not leaving your home. Otherwise it sucks.
the robot misinterpreted the sound as a sign that the man wanted to go to the bathroom. Later, every time a man cleared his throat before speaking, the robot changed directions and insisted to guide him to the restrooms.
And when he unzipped his pants, well, you know the rest...
if the man sneezes, then snorts, coughs, and finally clears his throat.
My neural net system indicates that there is a 75% chance he will spit next. In which case, he will want to use the sink in the bathroom.
___
It's the end of my comment as I know it and I feel fine.
permission from her landlord to raise a puppy for 18 months
permission from her employer to bring the puppy to work every day
mandatory attendance of weekly training courses
purchases of collar, leash, haltis
responsible for any medical costs under $500
daily socialization, including shopping malls, grocery stores, buses, doctor's and dentist's offices, theatres, movies, restaurants, elevators and sporting events.
(This is particularly challenging, since many store owners refuse to admit the dog, despite my sister's possession of a government-issued ID that explains the dog is a service dog. People often say, "But you're not blind!" -- they don't understand that people with hearing problems, as well as invisible disabilities, have these dogs, let alone that people need to train them first.)
Raising a puppy is serious work, but imagine you also need to teach that puppy to ignore food, stop at crosswalks, not chase sticks, ignore animals, and otherwise suppress many instincts. This requires an enormous amount of energy.
Those are just the requirements for the people raising dogs from 10 weeks through 18 months. For breeders and caretakers of puppies under 10 weeks, the people cannot work outside the home -- and they must take on many of the same challenges as the puppy raisers. As for dogs who finish basic training (at 18 months), many must move on to basic training with new trainers, who take on much the same role as the puppy raisers. Finally, after all of this work, the dog can be placed with a client (person with a disability) for specialized training.
Recruiting puppy breeders, raisers, and advanced trainers is a challenge for service dog societies, which also need to subsidize food, training and other products. And clients will eventually need to feed and care for the dogs. This is not a small amount of money.
Given all of these challenges, it would be interesting to see how a robot stacks up. If a robot cost $10,000 plus batteries, perhaps this is not actually much higher than the "value" of a dog that has been through 18+ months of training and must still be fed and cared for -- volunteer time and effects on the puppy raiser's workplace productivity should also be included in the calculation. Certainly, a robot may not provide companionship, but it may not be such a bad idea. With more time on their hands, volunteers could actually provide other programs for clients. Robots could be pre-programmed, so that the first 18-24 months of dog training could be skipped. And you don't need to buy kibble for a robot.
-- SYS 64738 --
As this topic is closely related to my own area of CS research, I was very interested to see the various comments on this matter. The first commenter (who pointed out that guide dogs provide more than simple directional guidance) has an excellent point; however, as others have pointed out, robots can process many types of information that dogs never could, can be programmed to do things that a dog simply cannot do, and can receive information from sources other than visual and auditory stimuli (as with the RFID tags).
As for the RFID tag conspiracy theorists: seriously, I completely understand your reservations about this technology, with all the talk of RFID-containing passports, and such. And I agree completely that this is a technology with a huge possibility of potential abuse, by agents as various as governmental agencies and "identity thieves". However, I think it's important to keep from losing sight of the fact that this is not an irreconcilably evil techology, by any means; it's not a bomb or a gun. There are so many ways this technology can be used to help people, as several users above have pointed out.
As for this robot itself (and its software): as we all know, speech recognition has a long way to go before it's really an optimised technology (especially when being used, say, in a crowded public place); nevertheless, I think the concept behind these robots is a great one, and I hope that he finds a way to make it marketable.
An assistance dog that has been attacked or in an otherwise threatening situation may be scarred for life and may not be able to continue as a guide/service dog.
Granted, training programs for assistance dogs may vary from place to place. However, when I checked a few websites (as well as that for the society for which my sister raises a puppy), I could not find any that say these dogs provide protection against intruders/attackers.
-- SYS 64738 --
I know this sounds cheesy, but the value of the companionship offered by a dog is something you haven't taken into account (and, indeed, can't have a dollar value assigned to it). I don't pretend to know anything about it, but I would imagine that having a disability such as visiom impairment would be socially isolating to a certain degree (most differences tend to have this effect in society, even those that don't limit mobility). The therapeutic value of having a pet is not to be ignored (a dog doesn't charge $100/hr :).
Another advantage is that a dog can adapt to new situations better than a typical robot. Also - though it has already been mentioned - a dog can help ensure safety (it is probably less likely to be stolen than an expensive electronic gadget)
This article caught my eye because I am a robotics researcher. Although I admire the intentions of the project, I think such a system design is hard to justify given other alternatives.
Here are some thoughts:
1) Using RFID tags for robot localization is inaccurate and probably not the best choice for fairly structured indoor environments like supermarkets. Just putting colored/patterned tiles on the floor or ceiling, or using indoor GPS would probably allow the robot to navigate better.
2) My gut feeling is that blind people want directions on how to find a particular product. Given simple audio advice similar to how GPS car navigation systems do: "turn left here, walk forward, etc" they would probably be able to use a cane or dog to get to the destination quicker, rather than having to follow around a slow-moving robot. So perhaps it would be better to give blind shoppers a small, portable indoor GPS enabled device when they enter the store that can always tell them where they are and give directions to where they want to go.
3) It would be very difficult to market something like this due to liability concerns. What happens if the robot accidentally leads the person into a sharp object, moving cart, irate grandmother, stairwell, etc... Any injury suffered by anyone involved that gets blamed on the robot will cause the personal injury lawyers to come knocking (especially here in the US, where the power of lawyers is very effective at discouraging new technology from entering society out of fear of liability).
4) As an aside, having RFID tags to identify various products that a blind shopper would want to buy, and then giving them a handheld scanner they could use to check products and prices on shelves might be useful.
-James