cancel
Showing results for 
Search instead for 
Did you mean: 

VR Gloves (my project)

CaliberMengsk
Explorer
Hey everyone,

I know I've had my vr glove project under my projects for a while, but recently I've been having more free time and actually got started working on it the last week or so. The idea is to use accelerometers, gyros, flex/bend sensors, potentiometers (volume knobs) and maybe a joystick, to make a highly accurate, non-laggy virtual reality glove. This first draft will be the proof of concept just to show what I mean. It uses and arduino leonardo (at least now it does), as well as the sensors mentioned above. At the moment, it's just writing all of the sensor data to a serial connection, then unity is reading that data (threaded so the game doesn't lag if it's to slow) and applying it in real time.

I wasn't sure if I should post it here yet, as I don't even have the elbow movement up and running quite yet. (Sensor is there, just I haven't built the elbow joint. This will be two plastic bars held together with a potentiometer to get how open the arm is.) but it's looking go so far, I think at least.

I still need to add the other two main fingers also, as well, the thumb has no input (first draft at least will be a joystick)

Let me know what you think! I'm hoping to get the elbow hooked up properly soon to be able to make a basic demo.

Here's the two videos I've made so far. Nothing spectacular, but still.
http://www.youtube.com/watch?v=--cOVZdpkWM - First video. Arduino Duemilinove (6 analog inputs), only one accelerometer and two flex sensors
http://www.youtube.com/watch?v=um3JUnARQQQ - Second video. Arudino Leonardo (12 analog inputs), two accelerometers, two flex sensors, and a potentiometer for elbow.

To Do List:
Craft elbow plastics and attach potentiometer and tilt sensor to it.
Add in other two flex sensors.
Create mount for thumbstick.
Add in thumbstick.
Make full demo project including picking up stuff as well as detailed actions (opening doors, windows, flushing toilet, things like that.)

EDIT:
All video links:
http://www.youtube.com/watch?v=--cOVZdpkWM
http://www.youtube.com/watch?v=um3JUnARQQQ
http://www.youtube.com/watch?v=c18tycYrDak
http://www.youtube.com/watch?v=282HJJANcio
http://www.youtube.com/watch?v=r15LuC7TD20
http://www.youtube.com/watch?v=dzMh7GttqXY
http://www.youtube.com/watch?v=DmXeFKzs-HQ
92 REPLIES 92

PatimPatam
Protege
Very nice! Quite impressive you got this far in just about a week.

Please keep posting your progress!

drash
Heroic Explorer
That's really awesome! Subscribed. 🙂

cybereality
Grand Champion
Looks cool. I'm still holding my breath for a glove.

CaliberMengsk
Explorer
😄 Maybe I'll just be the one that actually finishes it and has a decent one. 😄

ZeRax
Honored Guest
@CaliberMengsk, great work

Here is something i come up with under the STEM Kickstarter then i saw this http://youtu.be/oZXJYCuNtdo

STEM Controller mod
http://imgur.com/W61RaEL

CaliberMengsk
Explorer
The video you linked is cool. I don't think it'd work for something in a video game as far as the opening and closing, as it's lagging behind pretty far, and as everyone knows, in vr, low latency is key. Interesting use of the server to be a sensor as well as a haptic device. I may end up using something like that design in the end for stopping finger movement, but that'd be quite a bit down the line. My goal isn't the haptic feedback at the moment, but the speed and accuracy.

To be honest, I have a much better design in mind when it comes to accuracy over what I currently have. It'd take a much longer time to actually implement though, and would require something like a 3d printer, which I don't have. (believe it or not, this design would actually be considerably cheaper. It uses potentiometers instead of flex sensors. Each one of those flex sensors is $13, so when I have all four on, that glove is easily over $100. With the potentiometers instead, it'd be roughly a dollar per finger, so dramatic drop in price, but much more of an engineering problem, since to be accurate with them, you'd have to have a rigid glove.)

jbishop60
Honored Guest
Nice work. The next few years are going to see some real advancements in virtual reality.

Anonymous
Not applicable
CaliberMengsk: What about those online services where you send them a CAD file and they mail you the 3D printout?

CaliberMengsk
Explorer
The problem is, I'm not too advanced into cad design, so I don't really know how to do it where it would come out right. If I had the extra money, I'd buy a 3d printer... but I don't have much spare at the moment. T_T

My point is, to use those types of cad services are expensive if you are still building a prototype. You can find 3d printers for around $200 now if you don't mind having no heat bed. So after say 5-10 prints of tests, it'd be more worth the money to just get a 3d printer for myself and do it that way.

I'm probably going to do some kind of kickstarter after I get the proof of concept prototype done to give me funds for things like a 3d printer, proper cad software, more sensors, etc.

I also found a spiffy little chip on sparkfun that is both a gyro and an accelerometer in one, only costs $13, and communicates over an i2c (two wire) connection. That's cool cause then both the elbow and wrists' accelerometer and gyro would be on only two wires, instead of 12. Also, will drop the price from $80 in sensors to $16. (roughly $20 per accelerometer or gyro)

Only issue is, it's a QFC surface mount. I'd have to design some pcbs to get it to work right. That said, if I made my own arduino pcb with the right sensors on it, it'd reduce a lot of cost there as well. (1 arduino is over $20 on it's own when manufactured by others, a custom made one I did a while ago costs around $6 in parts to make.)

If all that pricing stays as I expect ($1 per finger for potentiometers, $26 for tilt sensors, $10 for custom pcb, a few dollars for the rigid glove and other various things), then this could be less then $50 per glove in materials, assuming I don't have to switch to a different micro controller to read the data. (It may be possible to find something that simply reads sensor day directly to a serial connection that may be cheaper then an arduino's chip, further reducing cost)

The cool thing about the new atmega chip is it has usb included in it. This way it's near the same cost as a normal atmega chip, but I get the bonus of it being all in one. As it stands, with the new chip, I can compress an arduino down to about the size of 1in x 1in and still have access to all of the pins. 😄 makes the glove very light that way (and extremely cheap when bought in bulk)