next topic »

[Reply #360 on:]

Oh yeah, it's been beaten over and over again on our quad forums.  Not only that, they've revived the horse, just to beat it dead again.  Put it on transfusions to repeat the whole process.  Not a week goes by that someone doesn't bring up some different way of asking the same question.  It's easy to think of, it's not easy to implement.  But it's still fun to dream.

[Reply #361 on:]

@cyberteq - You can get better than 10m, by waiting and averaging.  Certainly the relative position will improve in accuracy by doing so.  It is repeatable from one week to the next and very accurate.  The absolute position can be improved with WAAS.  LAAS will be <1m most of the time.  The satellite signal we use is the same signal the military uses, but their receiving equipment is different, and much more expensive.  There is no longer any encryption in the signal.  So I don't know what you're talking about.

@sbright33: Yep, you can certainly wait.  GPS signal takes anywhere from 10 seconds to 3 minutes to lock accurately.  Buy any GPS module and try it yourself, whether it's 4 channels or 12 channels, doesn't matter.  It will take time to lock, and you will notice it takes even longer for that lock to be within a 50 feet radius.  You don't see those quad in the video waiting that long, do you?

Any object will not, ever, stay in one single spot, dead on, with just GPS.  It will constantly move and adjust, both hover height as well as XY positioning.  You can augment a GPS with other sensors like I mentioned before, and you get better approximation.  However, you will not be able to have several of them fly like in the video without the external control.  By themselves, they will not be able to maintain that accuracy.  GPS is not that accurate.  Positioning is not only dependent on a GPS signal.  You need a compass, you need barometric pressure, both of which will change from location to location, as well as day to day, even hour per hour.

Till you try to build one of those vehicles yourself, and do your own testing, you won't understand it.  I have.  I've build 4 quads in the past 2 years, from small to large to carry a digital SLR.  There are hundreds upon hundreds of people who try, and fail, to get accurate positioning that is repeatable every single time.  It doesn't work.  You can calibrate your sensors in the morning and fly, then fly in the afternoon with the same exact settings and you'll notice it's off by several feet.  This is why we all perform autonomous flying with a 20x20 feet landing zone.  Because in the 10-15 minutes of flying that we do, it can change that much.

This is also why SparkFun opted to deny quad entry in their AVC contest from now on.  Seeing what they did last year it's just not safe.  People were calibrating in the morning, down to a few feet from the building, only to have the thing go haywire when it was actually time to fly in the afternoon.  It simply doesn't work.

And you are correct, the civilian and military signals are fundamentally exactly the same.  However, you forget that civilians only receive one signal, whereas military receives two signals at the same time.  This allows additional calculations, to account for atmospheric changes, so they are able to get a better location lock.  The clock signal received by the miliraty is also much accurate compared to the civilian clock signal.  This, again, helps with getting a more accurate position.

[Reply #362 on:]

I would have thought that there will always be some uncertainty with GPS, no matter how much you spend on the receiver. For example sometimes driving through the city my GPS tells me to "rejoin the main road" when I am ON the main road. Clearly the signal is bouncing off nearby buildings and throwing the calculations out.

Or even away from buildings, travelling through a cloud, or maybe on a hot day, might slightly change the time for the radio signal to reach the GPS receiver, and thus affect the calculations.

[Reply #363 on:]

There is, and all of those things you mentioned affect it.  Just the atmosphere will affect it.

[Reply #364 on:]

I got a open logic sniffer, probes and a generic 16x2 lcd red on black (could you believe I had not bought one yet?)

[Reply #365 on:]

I should also point that that receiving a GPS signal versus receiving a GPS lock are two very different things.  A GPS signal can be acquired in seconds.  In fact, your smart phone will capture anywhere between 1 to 20 signals in a very short amount of time (my Android does that), however the actual GPS lock doesn't happen for at least another minute, sometimes up to 3 minutes later.  Then it will constantly adjust, causing the apparent location to drift constantly by several feet.

Anyway, let's take this thread elsewhere and keep this forum for what it's intended, Your latest purchase.

Received today:
From Mouser:
ATAVRXPLAIN
10x N-Channel 60V/16A mosfets

From SparkFun:
2x 10K Trimpot
1x 100K Trimpot
2x 5-way Tactile Switch
A second Ethernet Shield
And another STK500 USB Programmer

[Reply #366 on:]

Hey GPS topic should start it's own new posting somewhere, perhaps sensor section, unless someone here just bought one and want's to share with us.

This thread is about spending money on cool stuff, stay on topic please.  

Lefty

[Reply #367 on:]

@kirash4 - I agree.  Sorry I didn't mean to start such a hot topic on the wrong thread.

[Reply #368 on:]

I got a GPS module from some outfit in China off ebay for something like four dollars, still haven't unwrapped it from stat bag.. No reason to think it's anything other than fine.  I got a great deal, but it's reasonably easy to find them for around ten bucks or so I would imagine (thinking dealextreme or whatever).

I was reading and waiting for Crossroads to chime in.. So there's obviously a reason.. So what is it?  I am talking about VORTAC, a radio beacon network that back in the stone age was used via signal triangulation off the frequency/tower chart to give pretty dang accurate positioning.  I am supposing it's a resolution issue to use actual VORTAC, but (I really don't know) would low power radio triangulation work within the range of reasonable open air lower power beacons?  GPS for large scale with finer granularity coming from triangulating off of local ground beacons of known position...

Got one of the elcheapo bluetooth serial modules, which arrived yesterday.  The seller says it is 5v tolerant though the module is labeled for 3.3v  .. Its got a backplane board of some type.  Anyone running these with TTL and 5v before I cook it?

[Reply #369 on:]

I got a open logic sniffer, probes and a generic 16x2 lcd red on black (could you believe I had not bought one yet?)

Now that you have one (red LCD), is it updating sluggishly? Every blue and red displays I saw have been sluggish. They're not extended temp models.

[Reply #370 on:]

no it seems to update rather quickly and cleanly for an LCD, I wish it had a better contrast though as the backlight leaks out everywhere so its more like red on slightly darker red but I think some tape will help that

[Reply #371 on:]

no it seems to update rather quickly and cleanly for an LCD, I wish it had a better contrast though as the backlight leaks out everywhere so its more like red on slightly darker red but I think some tape will help that

Astronomy and photography people like red. Fish tank people like blue. Both are inverted displays. The back lights are lighting the characters and leak out on the back ground (spaces with no characters).
The non-inverted displays use back light to light the background and characters are black. I like them better.

[Reply #372 on:]

I got a GPS module from some outfit in China off ebay for something like four dollars, still haven't unwrapped it from stat bag.. No reason to think it's anything other than fine.  I got a great deal, but it's reasonably easy to find them for around ten bucks or so I would imagine (thinking dealextreme or whatever).

I was reading and waiting for Crossroads to chime in.. So there's obviously a reason.. So what is it?  I am talking about VORTAC, a radio beacon network that back in the stone age was used via signal triangulation off the frequency/tower chart to give pretty dang accurate positioning.  I am supposing it's a resolution issue to use actual VORTAC, but (I really don't know) would low power radio triangulation work within the range of reasonable open air lower power beacons?  GPS for large scale with finer granularity coming from triangulating off of local ground beacons of known position...

Got one of the elcheapo bluetooth serial modules, which arrived yesterday.  The seller says it is 5v tolerant though the module is labeled for 3.3v  .. Its got a backplane board of some type.  Anyone running these with TTL and 5v before I cook it?

If you're willing to take this to a different thread, outside of this one, I'll be happy to continue the discussion.  Who knows, maybe together we will discover something new and exciting.  Maybe Robotics?  Or hell, it can stay in Bar Sports for all I care.  Just to make it fun.

[Reply #373 on:]

Astronomy and photography people like red. Fish tank people like blue. Both are inverted displays.

I've done all three. I like amber. 

[Reply #374 on:]

Created thread for GPS and positioning discussion: 

http://arduino.cc/forum/index.php/topic,91051.0.html