Meant to be Seen

The ABS based plastic that is conductive is brittle. I had to make many revisions to reinforce the design and get it to where it didn't crack under normal flexing of the fingers. It sits on top of a flexible PLC plastic which reinforces it and keeps it from getting bent to the ABS' breaking point. It will break if you can bend it to say 90 degrees or more, but if it's on the outside of the finger, it won't bend to the point where it breaks. I could possibly reinforce them more, but I don't see the need for it in my application.

This is the page that inspired me to do the 3d printed bend sensor route: http://www.kobakant.at/DIY/?p=4681

Good start ! But for precise, accurate gloves the goal is that all phalanx bones must be tracked... I'm a little sceptical about flex sensors... Durability, accuracy...

What's your use case? My use case is pinching/grabbing interaction within the VR space. I'm also ignoring the finger tips on my first try as I can't really see a use for having independent tracking of the joints closest to the finger tips. I am not tracking abduction between the fingers on this try because it's much more difficult to sew those sensors.

Don't get me wrong, maybe it's just my thoughts, but in VR you must see hand as you move it in real world for full immersion. Also pinching/grabbing/pointing will be much easy and accurate. Actually i don't know how hard it will be to make, but at least it must be your future goal.

I meant something like this, but for consumer market...

https://www.youtube.com/watch?v=yX-6vwRCMnI

I imagine having IMUs track each joint like the IGS Glove would be the superior option, but it's also expensive. Bend sensors are much cheaper and have been in use by the Cyberglove for decades, and bend sensors seem to work ok for them: https://www.youtube.com/watch?v=36BRjUHSAns

Any updates?

maybe you saw it already, but if not.. someone making what i mentioned here earlier (gloves with IMUs)
http://www.mtbs3d.com/phpBB/viewtopic.php?f=145&t=19590

But i still think that any VR tech needs RF position tracking, because every other things out there lacks accuracy or has problems of blind spots.

Recently, I spent most of my time on the glove. I designed a new circuit board for it. My first board was very noisy. The new one is perfect. I also replaced the 3d printed bend sensors with carbon paint on thin plastic strips. I can't really say if carbon paint sensors are better. I thought they were in initial testing, but now that I've spent more time with them, they don't seem to be. Maybe they're a little more reliable, but I made the switch because I thought the 3d printed sensors weren't being that responsive.

Yes, I saw the IGS glove with the IMUs to sense the fingers is coming to Kickstarter. I also saw your post about the 7 sensor glove versus the 16 sensor glove. I don't know if the cheaper glove is going to be good enough. The demonstration video looked good, but that was probably filmed with the more expensive gloves.

I agree that these relative tracking suits will need absolute positioning. I'm still working on getting multiple base stations up and running and was working on that earlier today.

Good to hear you still working on your project !

I don't know if the cheaper glove is going to be good enough. The demonstration video looked good, but that was probably filmed with the more expensive gloves.

No.. they stated, that in the youtube video was 7 sensor version of glove (cheapest one)

...

I recommend using some kind of vision system for your baseline. Should cover the range nicely and uses minimal equipment. Even if your baseline can't cover the entire range at once, you can adjust the baseline setup for discrete regions of the tracking volume. That's what I do with the robot; position the camera relative to the robot for given test ranges. It is nice to be able to do a continuous sweep, though.

What's the make and model of the robot you're using?

I had been using a sewing grid and manually moving the receiver to specific points, but yeah, I would love to just have an automated system at this point to do it all for me.

Make: Universal Robots, Model: UR5

It's on the cheaper end of 6-axis arms, but still very expensive. I want to say it retails around $30k, but can't remember where I heard that.

Something that might work well and be more cost effective would be to build a little 3-axis table that you can wheel around to different test locations. Could probably do a meter on each axis before it becomes totally unwieldly.

This is what I'm going with: http://www.openbuilds.com/builds/openbu ... chine.341/

I have the frame assembled and can wheel it manually at the moment with an area of 9ft x 9ft. Once I'm satisfied with it, I'll buy the motors and electronics to automate it. The receiver goes where the drill would go. I'm going to put the z-axis/lead screw on a separate frame that will move just the transmitter antenna array up and down.

Good call. I think that's the way I would go.

Krenzo,

I'm excited to read about what you're working on. I hope it's going well.

Have you seen the new DecaWave DW1000? Are you also using ToA like they are or TDoA? I am excited that you are reporting sub-centimeter accuracy (the DW1000 touts 10cm only) but I'm curious how you're achieving it. Do you attribute that hyper accuracy to the fact that your clocks are synced right now? Do you expect that accuracy to diminish when you convert to wireless clock sync in the future?

Have you read much about the 802.15.4a ranging specs and are you using it in your project? It describes a syncing technique using pseudo random pulses that is implemented in the DecaWave chip.

Either way, I truly hope you're closer to getting this thing out to a few folks to test via kickstarter or something. I have a wireless backpack PC with HMD setup I am testing with now. I've been testing IMU tags for full body tracking but without great results. If you don't mind could you update us on your progress? If you've stopped development on it or are hitting a wall, I may decide to get a couple samples of the DW1000. Let me know how you're doing.

Again, great job so far. A very impressive project!

-AC

angrycamel wrote:Krenzo,

I'm excited to read about what you're working on. I hope it's going well.

Have you seen the new DecaWave DW1000? Are you also using ToA like they are or TDoA? I am excited that you are reporting sub-centimeter accuracy (the DW1000 touts 10cm only) but I'm curious how you're achieving it. Do you attribute that hyper accuracy to the fact that your clocks are synced right now? Do you expect that accuracy to diminish when you convert to wireless clock sync in the future?

Thanks for your interest. I have seen the DW1000 and talked to Decawave briefly. It sounded like they wanted to adhere to the IEEE standard which does not emphasize sub-centimeter accuracy. I'm using ToA. I will move to TDoA when I make a wireless version. It does help that my clocks are synced, but it's not the reason why.

Have you read much about the 802.15.4a ranging specs and are you using it in your project? It describes a syncing technique using pseudo random pulses that is implemented in the DecaWave chip.

I'm not using any standards. Everything is wired together so I don't need to do anything like that to figure out what components are in the tracking environment.

Either way, I truly hope you're closer to getting this thing out to a few folks to test via kickstarter or something. I have a wireless backpack PC with HMD setup I am testing with now. I've been testing IMU tags for full body tracking but without great results. If you don't mind could you update us on your progress? If you've stopped development on it or are hitting a wall, I may decide to get a couple samples of the DW1000. Let me know how you're doing.

Again, great job so far. A very impressive project!

-AC

Since my last post, I've redesigned all of my circuit boards, assembled them, and I'm rewriting the software right now. I'm making good progress, and I quit my day job to go back to working on this full time. I've also formed a company: http://www.brenodi.com

Excellent. Very glad to hear it's still moving forward nad congrats on the big leap quitting the coding gigs. I know that takes guts.

At your pace, and from my research, you should be the first to market with sub-centimeter accuracy RTLS. The number of vendors producing chips for the 802.15.4* standard is growing (bespoon comes to mind) but the best accuracy any tout is 10 cm. Two French companies made EETimes top 60 startups. If you believe the press, UWB is poised for a comeback. Some of the new products marketed around the internet of things seems to imply that too.

Do you anticipate any added latency from additional tags in the field? How about as the number of base stations increases? I think I remember reading that you are doing one-way where each tag will be doing it's own passive locationing but I'm not sure of that. Of course if that's the case, and you plan to use multilateration, there will be an upper limit to the number of base stations before you start to add latency right?

Are you currently working within the FCC limitations and still achieving sub-centimeter accuracy?

If you travel outside of the 15-30 feet range offering the 1-3 mm accuracy, does the accuracy degrade quickly or can you still get within a couple centimeters at 40-50 feet?

Thanks,
-AC

...

Looks like it's making progress, though of course it would be nice to see a demo of the tracking.

Krenzo wrote:I've produced a new video to demonstrate my progress.
https://www.youtube.com/watch?v=NN97Qanfd_8

I'm getting the old 'This video is private'. Any public versions?

Sid wrote:

Krenzo wrote:I've produced a new video to demonstrate my progress.
https://www.youtube.com/watch?v=NN97Qanfd_8

I'm getting the old 'This video is private'. Any public versions?

Hi Sid, yes, my videos are now private.

I've signed an agreement with a company, The Void, to give them exclusive license to my tracking technology and work with them to further develop it. They will be creating large scale virtual reality gaming centers, and they've been the only company to contact me to not only share the same vision but have the funding and business experience to make it a reality. As per our agreement, they've asked me to remove all of my technical posts and take down my Youtube videos.

Congrats man!

Thanks, I'm really excited to be working on bringing this kind of very high level virtual reality experience to everyone and to now have the proper resources to accomplish it!

Yes, good deal. Another lesson taught here on MTBS3D. Talk is cheap. A lot of people yammer on and claim ownership of ideas, but the guys that get off their butts, learn something, and go out and build things are the ones that reap the most rewards.

Congratulations Krenzo, very well deserved.

Looking forward to trying one of these The Void systems in the future!

Glad to hear this Krenzo! Congrats!

Congratulations Krenzo! Well deserved!

That is pretty awesome news Krenzo, glad all your hard work is paying off!

Congrats Krenzo!

Congratulations! Great news!

But...
I'm a little concerned... Is this mean what we not going to see consumer related product in near future? Only on site attraction type stuff?

Correct, there will be no consumer products for the foreseeable future.

Krenzo wrote:http://www.roadtovr.com/the-void-is-a-v ... ironments/

That looks fantastic krenzo, very jealous.

Really exciting stuff I reaaaaaally hope a The Void center will open somewhere in Europe, or else, I guess I know where I'm going on a future vacation ;P Congratulations!