Find more videos like this on DIY Drones
For those that missed, it, today's Sparkfun event was a smash, literally and figuratively. It turns out that U(A)V's of all stripes have the same propensities as Charlie Brown kite. Aside from the Trees into which Jordi scored direct hits, the wheeled varietals made their own attempts to climb the Sparkfun forest.The wind certainly made itself known, and my first impression of the ardupilot (with pitot) was that it had very stable flight characteristics, wind notwithstanding, and that the waypoint acquisition was the low hanging fruit for improvement. I suspect the wind had decreased the accuracy outside the waypoint margin, resulting in several fly-arounds - but when the accuracy came withing range, it was quite effective in circling the building.
Comments
The mystery of the 12 second latency in the dynamic response of the EM406 GPS is solved:
It depends on which interface you use, NMEA or binary. If you use binary, there is a 12 second latency, then a 3 second "smoothing" transient.
If you use NMEA, there is a smoothing transient only. The NMEA interface to the EM406 provides dyanmic performance as good as the LOCOSYS or the ET312.
It is my speculation that "track smoothing" is turned on in the EM406 for the binary interface, and either it is always on, or I have not figured out how to turn it off.
Thank you very much for your help in figuring this out.
The bottom line is that the EM406 provides excellent performance.
Your video is clear. Your rudder responds very fast. There is a large gap between the 12 seconds that I measure for the latency, and your results. I don't know how to explain it.
I am going to file this under "unsolved mysteries" for now. I am retiring in July, I will have more time then to dig into it.
Bill
I just tested our "NorthEast only" code on the ground, and the rudder servo always responds within two or three seconds of me turning the plane. So I'm confused by your results. Here's a video of me demonstrating:
c_cc[VTIME] = 1;
c_cc[VMIN] = 1;
Sorry that I was not clear. All of my tests on the EM-406 were run in digital mode.
Clearly you have ample evidence that the EM-406 works well enough for you.
In fact, the EM-406 works well enough for my purposes as well, as long as the wind speed is not too great. However, a GPS with a shorter latency will continue to perform well in high winds.
http://deathpod3000.wordpress.com/
I thought his comments about IR vs sonar sensors were particularly illuminating, as well as how he dealt with loss of GPS sync.
Just to clarify, Sirf III digital mode is more than just baud rate. I'd definitely run your tests again in digital mode and see if it makes a difference. We do have real-world evidence that it works fine for our purposes (which is not say that there aren't better GPSs out there and that we won't switch to one of them soon; indeed, we'll probably be recommending the Ublox with our forthcoming shield)
I agree that the 12 second latency for the EM406 defies reason. I find it amazing that Chris did so well in the contest, with this latency. Yet the latency is there, I have measured it over and over again in ground testing. Paul Bizard has measured it too.
I am going to dig into the mystery some more, it bothers me.
In the meantime, if anyone has taken data logs that includes both GPS information and servo commands, the delay would show up as an apparent unresponsiveness of the plane to a change in servo position.
Unfortunately, there is no way to detect the delay be comparing GPS location with GPS course information. They will always be consistent.
It seems to defy reason that anyone could fly with this type of delay.
-
1
-
2
-
3
of 3 Next