So my friend put together a Motion Tracking system using RF. He'd been talking about the system to me for the last 2 years explaining to me what he's done so far. It's a very exciting project, although if I remember correctly he originally wanted to market it was a very cheap and very accurate replacement for MOCAP (MOtion CAPture) systems. Basically it uses an array of transmitters in each corner of the room to a RF receiver (What you would wear), with an on board gyroscope as well as a few other built in instruments to actively track it's own yawn, pitch and roll on top of it's XYZ coordinates. Although interested I didn't fully realize the potential until he posted this video.
[media]http://www.youtube.com/watch?v=mYyFUQbWC1E[/media]
For those who don't know the man in the video, he is Justin 'Krenzo' Krenz. The project lead and founder of the HL2 Mod 'Empires'. He's been off the radar since he left the team to pursue other interests a few years back.
What do you guys think? If you have specific questions as to how it works I'll either answer your questions with my existing knowledge of the system or ask him myself.
It's interesting but it would need a receiver on key body parts to be viable for mocap. Having to run wires is a downside.
Also, from his youtube comment: jkvr2013 2 weeks ago
g0band* is correct. To quote myself: "Target is several hundred dollars"
Seems like it won't be cheap.
[QUOTE=itchyflakes;40613692]It's interesting but it would need a receiver on key body parts to be viable for mocap. Having to run wires is a downside.[/QUOTE]
Yes, that's the point, for MOCAP solution you'd wear a 'suit' that simply has a few RF antenna's on key points of your body hooked up to a light 'vest' that does the heavy computational lifting.
[QUOTE=itchyflakes;40613692]
Also, from his youtube comment: jkvr2013 2 weeks ago
g0band* is correct. To quote myself: "Target is several hundred dollars"
Seems like it won't be cheap.[/QUOTE]
of course not, traditional optical systems cost tens to hundreds of thousands of dollars. This would allow me to do the same thing big developers and Hollywood studios do with vastly more expensive systems, out of my basement, for around $500.
Now I from what I understand, if he were to have custom IC (integrated Circuit) chips made to cover the functions of a few of the specialized chips he has on the sensors and transmitters that would bring costs down by a hundred or so but I understand that's still not super cheap.
The problem is though that for 1/1 mocap you can't use "a few" antennas
[QUOTE=LordCrypto;40642036]The problem is though that for 1/1 mocap you can't use "a few" antennas[/QUOTE]
isn't 3 enough?
[QUOTE=Shadaez;40693663]isn't 3 enough?[/QUOTE]
Motion Capture for movies and games involves tens to possible even hundreds of points on the actor's body which is why optical tracking works great. Radio waves are used for long distance surveying and measuring, but I can't see it working well for tracking people and limb motion. Would be useful for controllers and peripherals though.
It's really similar to what Google does with location estimation with WiFi, but with this kind of system, you could use a signal that can penetrate a lot of different materials way better, WiFi tends to bounce off big bags of water, like humans.
If you managed to get the receiver-size down a lot, and used an unlicensed band, and range wasn't a problem, you could use this for in-doors GPS at large places, like malls etc., since the locations of the transmitters could just be inputted in by whoever owns the mall.
And Google is pushing big with indoors maps for Google Maps, they did a talk about it at Google I/O this year. But then again, WiFi is already built into the phones, and it's good enough.
All RF bounces off people. There are companies working on large indoor tracking using standard wifi signals. I was contacted by one such company after posting this video. My system is not good for that application because it favors precision over range because I want to do tracking for virtual reality.
[QUOTE=nikomo;40694812]It's really similar to what Google does with location estimation with WiFi, but with this kind of system, you could use a signal that can penetrate a lot of different materials way better, WiFi tends to bounce off big bags of water, like humans.
If you managed to get the receiver-size down a lot, and used an unlicensed band, and range wasn't a problem, you could use this for in-doors GPS at large places, like malls etc., since the locations of the transmitters could just be inputted in by whoever owns the mall.
And Google is pushing big with indoors maps for Google Maps, they did a talk about it at Google I/O this year. But then again, WiFi is already built into the phones, and it's good enough.[/QUOTE]
but that would require a separate app for that, whereas google is already rearing to go with indoor maps
[QUOTE=LordCrypto;40697190]but that would require a separate app for that, whereas google is already rearing to go with indoor maps[/QUOTE]
It would be a feature for google maps
[QUOTE=Bonzai11;40720147]It would be a feature for google maps[/QUOTE]
imagine 50 different companies all making slightly different indoor rf positioning systems
is google really going to make deals with all of them?
What an amazing solution! With a little miniaturisation it looks like we might have found the missing piece of the puzzle for a new game concept I'm developing with some friends!
A few questions:
1. How well does it travel through walls (i.e. Could one set of transmitters be used to cover a large subdivided space.)
2. Does the RF used affect other radio signals in the area? (Could it be used successfully along side a wireless network.)
Regards,
Simon
Sorry, you need to Log In to post a reply to this thread.
