I recently acquired the Dexter Industries dGPS sensor to be used in a robot designed to compete in the SRS RoboMagellan contest. I have a few questions about it.

1. How fast can it be polled? I see in Xander's sample it polls the data ever 0.5 seconds. Can it go any faster? Is it know, or do I just have to try?
2. How accurate is it? Comparable to a phone, say? I'm not sure if I'll need to incorporate encoder data to supplement the GPS data for position.
3. Once it gets a workable signal, how apt is it to lose it? I've been trying it in a pretty enclosed area, but it seems to work fairly well.

No solid answers here, just some related rule-of-thumb information.

In general, GPS modules are accurate to within about 5 meters (unless they are a DGPS, which is something totally different from a dGPS). Typically a GPS will update 1, 5, or 10 times a second.

I have no idea what the specs are, for the GPS on the dGPS, and I don't know how often the dGPS FW allows for updates. I'm guessing that it's probably accurate to within about 3 to 5 meters. Whether or not you need to use encoder data as well will depend on the setup/project/contest.

Once a GPS module locks, it should be able to keep that lock, because it will know which satellites it's communicating with and where the other ones are.
According to the dGPS wiki (http://dexterindustries.com/manual/dgps-2/), you can get an accuracy of 3 meters. But since it's just a commercial module, it is really only guaranteed a 5-meter accuracy. When people talk about higher accuracies, they're usually talking about a GPS with augmented accuracy (WAAS, DGPS <-that one really confused me!). DGPS=differential GPS, not Dexter. Apparently you can get 10cm accuracy with that? But RoboMagellan doesn't seem to allow DGPS.

It seems I *can* cut the polling down to 1/4th of a second, but it doesn't seem to be updating much faster anyway. I do live in a pretty enclosed space however, so in the field it should perform better.

The accuracy should be good enough for my specs. And yeah, dGPS and DGPS, big difference! It sure confused Google.

I did some tests today. The GPS seems to be working reasonably well. It's just MY code that was failing. I need to try something new... it seemed to be actively trying to avoid the waypoints...

And did I mention, this needs to be semi-working by August 3rd

I must be tired. I never use this many smilies...

The GPS has a max frequency of 4Hz, that's the spec. It can go to 1Hz as well, I believe, but it's been a while since I last looked at the actual specs of the chip.

= Xander