Design of Local Area GPS
I'm having some troubles finding a reliable and accurate DGPS ou GPS module for an indoor application, and I wonder if someone can help me.
I'm looking for modules capable of giving an accuracy close to 0,15m, but I'm not having much luck.
Thanks for all.
André
Indoor, when you not can receive direct signal from satellites, must received signal bounce a number of times in walls/floor so total distance to satellite easy becomes several meters longer then for direct signal. DGPS is a service, not a hardware or something that can be received from GPS satellites. This kind of service do seldom offer better accuracy then +/-1 meter. With statistical methods can it sometimes be somewhat improved.
"when you not can receive direct signal from satellites, must received signal bounce a number of times in walls/floor so total distance to satellite easy becomes several meters longer"
I've heard about pseudolite, and it seemed to resolve some of that problems due to signal reflections on the walls since the signal comes from very long distances. But I don't know if I can find some device that can play the role of gps satellite but in a close distance allowing to measure the coordinate system.
Thanks once more.
DPGS only works in a cleared area, not indoor. The pseudolit idea is principle solution, but I guess, you are underestimating the complexity of GPS transmitter systems. For local area position determination, I would look for short distance RF or optical time of flight systems, e.g. those systems that have been discussed for football or tennis referee support.
you can not do a 15cm accuracy using RSSI indication. You need a more accurate system. There area a couple that can be used. I design such things
Rich
"For local area position determination, I would look for short distance RF or optical time of flight systems, e.g. those systems that have been discussed for football or tennis referee support."
You're right, but still the problem is that RF based systems have. Those problems are related to the frequency necessary to acchieve those specifications, the frequency needed is in the order of 60GHz, could be less if phase identification is used. Still I don't know if it works.
---------- Post added at 14:02 ---------- Previous post was at 13:45 ----------
"You need a more accurate system. There area a couple that can be used. I design such things."
Rich, those systems are hybrid I would guess right?
No, why. The said ball locator projects have been using UHF or 2.4 GHz frequencies and achieved better accuracy than 15 cm.
Didn't say it's simple stuff. I just say, it's feasible and has been done by others before. It's beyond a simple project and requires profound RF engineering knowledge.
Thanks for all.
---------- Post added at 14:52 ---------- Previous post was at 14:51 ----------
Thanks for all. Do you have some literature you think would be helpfull?
Do you want to have a look at infra red or laser rangers. They easy give a very high accuracy when used for range metering or obstacle avoidance system for mobile robots. The only disadvantage is that they are short range, i.e up to maybe 10 meters for infra-red, laser range meters can cover longer distance.
Those are not giving the absolute position, but just the distance to obstacles. I know that this is not the same, but may work for you. Certainly it work for the mobile robots.
I'm open to every solution, the only problem of lasers is that you'll need clear view of the point, and I'm trying to gather intel, about systems that provide not only coordinates but a reference, but I'm not having much luck until now.
Well, GPS make sense when it has open sky and your absolute accuracy is up to a few meters. Sometimes the GPS unit is making a correction, based on speed and using additional inertia sensors and etc. by this the accuracy improves. But for indoor and with distances of 15 cm it is not going to work - this is at least what I know, maybe there is some improvement in GPS which I am not aware of and it do the job ...
The industrial robots, also the commercial are using different approach :
1. white line on the floor, similar to cable in the floor that makes magnetic field - then the robot just follow the white line or the magnetic field and is moving in a kind of track
2. IR or RF GHz range base stations that provide a kind of beam orientation - when you robot see the base station it knows where it is because each station transmit a very narrow beam and also its code, then it is up to the software to make the orientation and to navigate ...
3. IR or radio fence - this is the same as above, but is used to make an artificial obstacle - when the robot see this barier it goes around or turns back
4. laser ranger - this could be a laser scanner, it scans the room and know where are the obstacles, then it moves in and scan again, it is complex and may require complex software, the hw for the scanner is also not simple, but there are ready solutions which you can buy. I have seen a few in internet, they are really cool for robots :) .
5. laser dot orientation - you beam many small lasers, which one is transmitting its code, when your robot see the laser spot it will know where it is because it will compare the laser unique code with the map that it has inside. Making such a grid of lasers is not complex, but will create an environment that is not really human, at least I don't want such infrastructure at my home :)
Hope this gives some ideas to you. Don't hesitate to post again, and please, describe your application, what is your goal ...
It would be much easier to suggest a solution, if you tell a few details about the application problem.
The goal is to develop a system capable to resolve coordinates of a given position to optimize objects allocation, but I'm having troubles with the resolution of the system needed, and with the reference system yet to be chosen. Hope this helps. Thanks for all!
My problem now is to define the reference system, because of the spaces with obstacles in the field of view.
Obstacles will affect both optical and RF systems, the latter in a different way depending on wavelength, obstacle size, shape and electromagnetical properties. Overall distance also plays a role.
Cortez, I have a similar requirement. Have you looked at Symeo LPR products, they were advertised on your thread. Do they satisfy your requirements of accuracy, handling reflections, etc?