Forums > MaxMSP

Getting position of persons

November 29, 2012 | 10:53 pm

Hello Everybody!

I’m looking for the best way of recognizing the position of the persons in a big room for installations. Making this with jitter I know there’s no problem. However, the idea is that each person should have something like a serial number for tracking his position wherever he would be in the room. What do you think is the best way for communicate all this data into MAx? And of course, which do you think is not the most expensive way, for example for 30 persons.

– Jitter: I think it’s not possible. Maybe we could put a color or a number, but if this person goes out of the room, when he comes back to the room, Jitter we’ll assign another number or color.
– Sensor GPS: giving an small arduino constructed with GPS sending by Xbee the position to the "father" of arduinos, which will be conected to Max.
– Radiofrequency Identification (ttp://www.phidgets.com/)
– May be it’s possible to connect all mobile phones of the persons to a wifi system that sends and receive information from four routers and gets x and y position.
– IR cameras?
– Others?

So complex to get the serial number of each person. On the second hand, thinking about giving an abject like a small arduino, this could send data to max like "device 1 pos.x 14 pos.y 400". Just tracking the x,y (and z) position, it could be great.

What do you think about? Which could be the most eficient for max? And also for the economy?

Thank you very much for your ideas.


November 30, 2012 | 10:23 am

Kinect? http://cycling74.com/forums/topic.php?id=37285


November 30, 2012 | 8:34 pm

Yes, I saw it, very cool project! However, my question is more for get the position to send to each person-position new data for them. I mean, the question is to receive the positions of all people for sending to them new data in function by them position. Thanks a lot.


December 1, 2012 | 9:24 am

How many people do you want to track? Can you give them something that helps identify them? (device, a hat, a wristband)

If you don’t do it with analysing camera information (is suppose through jitter) you can do what GPS does: have 2 beacons sending out timecodes and each person has a receiving device, calculate their distance from the delay in timecode. Arduino won’t be fast enough for the interpretation though.

You can also do the other way round: everybody carries a device that sends out timecode with I’d string and you capture these with 2 or 3 sensors and calculate their position from the time the signal arrives at each sensor.

Both you’ll probably need to find some dedicated hardware, not sure if anything exists.

If you have, say, max 8 people at the same time I would think about having some kind of visual identifier – it could be infrared too. Which mean you won’t need much more than cameras and jitter.

Final idea: how about each person has a pulsating light on them ( again, could be infrared) but they each pulse with a different frequency (or even pattern) jitter tracks the position of the light sources and uses the frequency or pattern to identify who is who. Just need to find a way of making the light visible from all corners.


Viewing 4 posts - 1 through 4 (of 4 total)