Picture of eyeRobot - The Robotic White Cane
Using the iRobot Roomba Create, I have prototyped a device called eyeRobot. It will guide blind and visually impaired users through cluttered and populated environments by using the Roomba as a base to marry the simplicity of the traditional white cane with the instincts of a seeing-eye dog. The user indicates his/her desired motion by intuitively pushing on and twisting the handle. The robot takes this information and finds a clear path down a hallway or across a room, using sonar to steer the user in a suitable direction around static and dynamic obstacles. The user then follows behind the robot as it guides the user in the desired direction by the noticeable force felt through the handle. This robotic option requires little training: push to go, pull to stop, twist to turn. The foresight the rangefinders provide is similar to a seeing eye dog, and is a considerable advantage over the constant trial and error that marks the use of the white cane. Yet eyeRobot still provides a much cheaper alternative than guide dogs, which cost over $12,000 and are useful for only 5 years, while the prototype was built for well under $400. It is also a relatively simple machine, requiring a few inexpensive sensors, various potentiometers, some hardware, and of course, a Roomba Create.

Step 1: Video Demonstration

Picture of Video Demonstration
1-40 of 52Next »

I see my white k-9 inspred the guys in the back room. ozi

wow i spent tyme reading all the good the bad and the ugle posts. here keep up the good work.
Personally i believe that for that single purpose, yes, it might be better than a seeing eye dog. But, let us assume that this blind person is natvigating New York City for a few examples of why this might not be quite as good. Seeing eye dogs are trained to keep their owners safe, not just from tripping, but from walking into a street and getting run over, etc. Dogs also have an acute sense of distress, as in, they know when something is wrong. I've heard countless stories of regular dogs saving their owners from fires, finding help for someone having a heart attack, saving drowning people, etc. Speaking of water... that doesn't quite look waterproof, so if it's raining, your screwed. Again with the New york theme, someone walking around with that is probably 100 times or more likely to be mugged than someone with a dog, and they lose the protection provided by a dog. It's a good effort, but i think if i go blind, i'll stick with the dog.
Waterproofing isn't really a problem. I can usually waterproof a robot in a few minutes to a few hours. I also am afraid of dogs so I would rather use this than a dog especially if I was blind.
try a trianed eye seeing quarter horse. an yes thier allowed ever wher that a dog is. if thier not that place wont be there 4 long!!!
eyerobot5 years ago
Needless to say, I love the name!
I have been working on a handheld device, That gives audio feedback according to distance, And when its done I may actually be able to fit it in old television remotes.
Of course its not as elaborate as this project, And requires people to train themselves to recognize what the sounds mean, But at least there are some guinea pigs(neighbors) close by, That are visually impaired.

But I really like this idea, Good job. 
to power the device look up thermal electric generator online. that way the user can power the device from thier own body heat.
That looks good for flat ground but I am visually impaired and no pun intended I can see a hole in your cane, holes and ledges.

Can it tell a curb from a wall and would the cane let you walk into a hole, I can see the cane running along the top of a curb and letting the user accidently step off the curb and fall into traffic.

I can find my way between the telephone poles it is the ground and all the landmines left by sighted people that get me.
Timmers2 years ago
Have you seen this, could be a really cool creation.

Sorry for the very late reply, I try not to "reopen" conversations that are a year or more old generally.

Mostly I wanted to add my own "me too" comment though in response to your CAPTCHAs remark. CAPTCHAs are very annoying to the sighted as well. That being obvious or not to the other group of users, I wish there were better "are you human" checks that were not so annoying. In fact that is the reason that this reply is at the top level and not a direct reply to yours, the reCAPTCHA JavaScript used to insert itself into this pages DOM when replying ( but not "New" comments ) is faulty. After checking it was not a Google Chrome bug by trying it in Firefox and Safari as well I concluded I had done my due diligence and gave up on further tracking down the bug, leaving it to the developers responsible for instructibles.com to follow up on. Maybe someday they will be a thing of the past, we can only Hope :)

And secondly to thank you for your ( even if it is minimal ) insight, it really is hard to determine things like this for a group or target audience of any kind even if you are specifically trying to and have the best intentions. Thanks :)
There's a reason why the white cane has remained largely unchanged since it was invented something like 70 years ago: It's simple, cheap, and effective. :)

Actually, it has changed in that time-frame. The first blind guy to use a "white cane" was Jacobus tenBroek, a blind civil rights attorney at a time when blind people were street musicians and broom makers working for sub-minimum wages in government sweatshops—er, I mean "sheltered workshops" (this practice continues as of December 2012 in the United States, with your friendly profiteers at "Goodwill" leading the way paying as little as 72 cents an hour, but that's a whole separate discussion…)

*ahem* Anyway, the guy who invented the cane originally painted it white, and put a red tip on it so that sighted people wouldn't trip over it. Being blind as he was, he did not realize immediately that the red tip was both unnecessary and indeed not helpful for the purpose. As it happens, being able to see it doesn't matter if a person isn't really paying attention. ;) You could light it up with neon and some people would still trip over it.

As always, the "professionals" got into the act and began trying to prescribe how long a cane should be, and complex maneuvers for how one should manipulate it, etc. These are the same brilliant folks who wrote manuals teaching a 12 step process to switch from holding on to a person's right arm to holding on to their left. Yes, really, and the procedure isn't one I'm going to perform on anyone who isn't female, in her 20s, and cute. They also wrote a whole manual to teach a blind person to take a shower.

The blind themselves… continued to evolve the concept. The cane got lighter—much lighter! And longer. How long? Well, a good rule of thumb is to put your back to a wall, walk several paces away, turn and walk right into the wall at a good pace while using your cane. When your cane hits the wall, try to stop before you do. If you wind up hugging the wall, you need a longer cane. ;)

As noted, the early cane was designed for maximum visibility to the sighted world. Blind people today don't generally regard this as important. The cane is a means for us to get around, nothing more. The idea that sighted people would or should watch out for us because we're blind just doesn't mesh with the real world. We are able to learn to watch out for ourselves, and if we don't who will?

Of course, just to really throw a wrench into the works for figuring out your target audience, what I describe above is just one approach. It's the one approach that works and makes sense, but it's not the only one in use!

A lot of blind people don't obviously start out that way, for example. They tend to lose vision later in life, at which point vocational rehabilitation has little to offer them, arguing that they have little to offer the workforce if trained. Many in the vocational rehabilitation industry are sighted people who used to write those manuals on showering I mentioned, and they basically haven't got much faith in the ability of even those who have a lifetime of experience in adapting to blindness to be otherwise normal, functional human beings. Still others reject the independent-minded thinking I describe above, either for political reasons or because they just don't want to. The latter of these often believe the world should adapt to them, rather than they to the world. IMO it's not very realistic.

There are some out there for whom this device is probably very interesting. It just isn't going to fit into the budget for most of us, no matter how cool it might be. My canes cost me $25-40, depending. I'm using the same one today I picked up back in 2009, and it's stood up to every time I've dropped it, had it stepped on when I laid it down for a moment, or other form of abuse. A good car door slam will crush or shatter it, but short of that if I take care of it, it should take care of me. :)
modmadmike4 years ago
To all those who think that it will be completely useless because of stairs/bad terrain/etc, this is a prototype, and the final version can easily be made to conquer stairs by being built without the iRobot, and instead using motorized wheels therefore needing to be simply moved by the user like a regular cane. Or it can be built to use audio feedback instead of force therefore dropping the need for wheels entirely.
jjaydam4 years ago
Great Job
I don’t really see how this present design can navigate things like stairs (or even curbs), something that would be an absolute necessity. Also, you’re going to have to think about terrain that a Roomba cannot traverse. That’s important because we must walk through mud, gravel, snow, etc.

It seems to me that you are approaching this with the notion that a person using a cane must bumble along in the hopes of finding a clear path, whereas a service animal sees and thinks about an obstruction-free route from origin to destination. With respect, that’s sighted-people-thinking. We blind folks don’t do that, and in fact we don’t want to do that.

I’ll do my best to explain why.

A cane user moves from origin to destination mainly by moving in straight lines, going from landmark to landmark (or static obstacle to static obstacle, in your way of thinking). Static obstacles that are familiar to us help us know where we are in relation to the world around us. Static obstacles that we do not know serve as waypoints to get back to wherever we were before we decided to go wherever we are going. We will tend to take the same route between two places until we become more familiar with an area to reduce the number of surprises.

When you encounter a dynamic obstacle with a cane (ie, a person or other thing that won’t be there later), you simply go around it and continue on your way.

Working with service animals is a little different. The dog will tend to take you around all obstacles automatically, unless you slow down and approach an obstacle to identify it. It’s a little different in appearance to an uninitiated sighted observer, but the navigation by waypoints is fundamentally the same. It’s just that the cane user will "run right into something" (that is, find it with their cane, often intentionally) before going around or whatever is necessary, whereas the dog will indicate to the user the presence of something in the path by leading the user around it.

The person using the cane or dog must still know how to find their way from place to place, cross streets, recognize and navigate hazards, etc. The dog doesn’t "see for you" in any real sense. But most dog users do walk faster than most cane users. Personally, I prefer the cane, but I'm not an animal person. The reason nobody has managed to replace the $35 cane with high-tech solutions yet is that the high-tech solutions don’t work everywhere a basic cane or a well-trained dog will. Doesn’t mean it can’t be done, just that the people trying thus far have not been thinking about how many environments blind people traverse and what exactly they expect the replacement to do for them.

And um, for those wondering, I’ve been using talking computers since 1982. My first was an Apple ][e with an Echo synthesizer. In the past 30 years, we’ve managed a FEW advances in the state of the art. ;) Undescribed images are a problem for those with little/no residual vision, and CAPTCHAs are just annoying, but otherwise we can usually manage.
ValleyWolf5 years ago
I'm a blind guidedog user, and I think this is definitely the next generation of service animals. The only true flaw I can think of however, are the personality and expressions which come with the living animal. I was thinking to compact the size that one could construct the electronics(sensors, batteries, etc.) up the length of the cane, leaving the motors and, perhaps narrower and taller, wheels at the base/point. I believe there are many versions of this project worldwide, so good luck in that race.
Did you just say you are blind?
Yes I am blind, although I said it two months ago. Wy?
How do you type or know what I've typed then?
BUMP keyboard and bump screen to read
And the pictures?
But their blind, they can't see at all
ok....... i know that, however, many leagaly blind people i know can see, but faintly and can only make out things close to them.....
i have a friend ho is mostly blind, he an see a little, he and his wife use jawz, i know because my dad has to fix it about once a month
I believe that is a little bit of a strange statement "how do you know what I typed".
Many people with a significant visual impairment have some degree of residual vision. Hence to be blind doesn't mean pitch darkness. It often means a slight amount of vision remaining that is not enough to function.
Hence the visually impaired can use things like magnifiers, screen readers (It reads out the text on screen), self voicing for typing (Windows 7 is using this).

I love ressurecting threads for justice!
this was on xp, before vista even.............
elbeil4 years ago
Having a visually impaired daughter, I can strongly encourage you to continue to work on developing a consumer friendly product. Any input I can add from the standpoint of the visually impaired will be gladly given.
aeray4 years ago
A great Kickstarter project for sure. Keep it up.
Ghost Wolf6 years ago
what about crossing the sidewalk?
thats a really neat idea, and it could be really useful and could help alot of people..... KEEP ON TRUKIN
sedition7 years ago
Awesome Instructable. Great seeing that you were testing it in a way that would be indicative of it's use, i.e. blind-fold. So many great ideas are hampered (or problems even compounded) when those without disabilities try to assist those with. I'm taking the same approach with the Linux distro I'm throwing together and I'm amazed at how frustrating it can be when you run into a problem that is easily solved with raising that blind-fold. Unfortunately, those pit-falls aren't overcome by the people you are developing for, but at least you can account for and come up with solutions to those issues before your users run into them... Next step(s) getting rid of the Create? Could you not just use the Arduino as the brains and just attach your own motors, power source, etc.? It would reduce the weight, cost, etc.
Whats your linux distro for?
A good name7 years ago
It's a seeing eye dog and a cane.... I saw this in a magazine called robot... pretty cool.
grayecastle7 years ago
what is it???
juliofo7 years ago
Unbeliable good idea and a very good job! certainly a plus plus Congratulations from spain. Keep going. Saludos
static7 years ago
I as well wondered about stairs and steps. Can it be programmed to remember routes? I believe the blind count their steps from on point to another to know when they have to make a turn down another hallway or doorway. Have you yet approached your cane using fellow students to get their input product evaluation? I applaud your effort, but it's nearly impossible for an abled bodied person to perceive the needs of disabled. The disabled need people with your initiative, keep it up and involved the disabled one on one as well if you haven't yet. We can all learn from each other.
AMalePoet7 years ago
Good ground work. I too see potential in this. When the handle is dropped a deadmans switch? Much of the tech stuff is beyond me with out actually seeing it in front of me. My input is more practicle ergonomic fare. A cane is a round stick so is a one for the blind. The stick used looks like a street hockey stick. Perhaps a graphite tent pole with shock coarding inside to aid in portability. Also a smaller gauge stick would fit a brouder range of users. In future versions consider designes that may want to tollerate moisture or at least avoid deep puddles. Larger wheels would also make irregular terain easier to navigate.
SuperTails8 years ago
That's a wonderful idea but how would you know if you're home or in a place you want to be in. The advantage of a seeing eye dog is that a dog barks to let you know if your home. It also lets you know by stopping whens theirs traffic. All ideas have to start somewhere and this is a great start but it should be have some extra features on it before it becomes a seeing eye dog substitute. I would hate to be a blind person getting hit by on coming cars because I didn't know when to start walking or where the other side of the street is. Anyway this is a great start and good job.
This is a great idea, but i have a few questions. I have worked with sonar and I have also read many papers on them. I have noticed that sonar sensors sometimes give different readings based on density, i.e. a foam or soft substance that will absorb sound often gives a false reading to the point that some really porous, low density objects are all but invisible to the sonar. That brings me to the real question here, have you encountered any such sonar problems, and if so how are you combating them. The second question that I have is more of a curiosity. It looks like you don't have any forward facing blind spots, but how does backing up/turning around work, if at all? Thirdly, If I read this correctly, you just take the readings from all of the sonar at once and then calculate a vector that aims towards the middle. If that is the case, what happens when the eyerobot is directed into a narrow dead end? Thanks, once again this is a great idea!
1-40 of 52Next »