Apologies if this has been asked before, I am interested in some of the 'cooler' things that could be done with Microcontrollers and processors. Anyways my question is two fold, so bear with me. I am not really looking at doing anything like this (way beyond my dreams!) but was interested.
The first question is in regards to internal building locations. I see all of these cool tracking and location projects but i wonder if there is a way that the GPS location can be determined to a recent level of accuracy. 2m or 5m may be useful for outdoor environments, but when you are looking at indoor environments (where you can recieve a GPS signal) if you want to be able to determine where you 'are' in a building (assuming you can map plan to GPS coordinates) you would at most only potentially be able to get away with an accuracy of less than 1m (which seems impossible based on current 'normal - non military' systems.
The second question, which is sort of relating to the first is; There seems to be the idea of really cool HUD systems for Army, Firefighting and so on, where people can be walking around 'in a building' and be able to see the relative location of each of their people in their groups. I presume each person is wearing a transponder which share's their relative locations, including which direction they are in (and distance), so when the user turns their Heads Up Display in the direction their team is in, a symbol of the team shows up with their distance.
What i wonder is, other than a direction (for example bearing) which could be communicated to the team, so they know person is at your NNW bearing show up when you look that way, how can the distance be calculated, when techniques like GPS would have insufficient accuracy in an internal environment. If the accuracy of GPS was good, i would understand that, you could get bearing, distance, lat, long, elevation etc from that and it would be useful because you could compare that to your other team, but that would not be accurate enough.
Sorry for the bad way of explaining it, but wondering how people can determine the relative distance and location of different 'arduino' units in an internal building when something like GPS would not be sufficient, and work out the distance.
people can be walking around 'in a building' and be able to see the relative location of each of their people in their groups.
Off the top of my head I can't think of any way to do this, GPS is not accurate enough and won't work inside anyway. With any form of radio you either need several "master" nodes that can get direction from transmitters and then triangulate, or each person's gadget has to be able to determine direction and signal strength of it's neighbours. And given that the signal strength would be a bad indication of distance due to walls etc I don't see that working.
The last technique is actually used quite well by those tiny swarm robots, but that's in in a very controlled environment.
I suppose each person could have an INS and that data could be transmitted to all other units. How much that would creep over time without a reference I don't know.
That is what really got me interested, from a logic perspective you see people doing this so there must be some way to do it, but it cant really be GPS since it does not have the right level of accuracy to be able to see it. My thought was that it would have to be something like compass and so on, at least then you could see which direction your team was in, but would not really be able to tell distance.
Triangulation would not really work, because in all of these scenarios the idea would be for it just to work, use the police SWAT style team as an example - chances are they are not going to have time to set up towers or beacons around the place, then do a raid. They just want to get in their and get it done. I guess you could look at direction AND signal strength, but then that does not really tell you how far away someone is since walls, and so on within an internal environment would potentially affect the measurements.
Id appreciate any other thoughts, it is going to really annoy me not knowing how it could be done (strange i know, but still interested)
It may just be a pipe dream or concept, but this is the application i saw it from and was wondering how they managed to get the breadcrumb trail (where the person had walked), and where their team and other vehicles were:
That's pretty neat, well above an Arduino's capability I think.
I don't think it showed where the other guys are located. From what I could see it showed what they were seeing so I assume they all have head cams and the video can be routed to other headsets.
It did show an exit path, I would say that was using INS to record the way in then just reversing it to get out.
Nothing in that video said "available for use today". The technology, if/when developed, will likely require just a bit more processing power than an Arduino can provide.
The breadcrumb trail could be created using accelerometer and gyro capabilities available today. It is nothing more than a "you got here this way" track.
No worries, i figured it was way above the capabilities of the Arduino, just was really interested in how it could be achieved (in general). Can you point me in the direction of any IMS information? I am curious about how something like that could be used to map where you walked in a building (rather than GPS).