For my knowledge of GPS sensors (for some reason I had
to use many type of such sensors) that:
1) the position accuracy of a 'standard' GPS (i.e. non
RTK or dual frequency ones) is never better than 1-2mt
and such performance can only be achieved with a good
satellites constellation in sight (IMHO this
performance is reached in less than 20% of the working
time);
In standard condition (60% of the working time) the
accuracy falls to 3-5mt;
2) ground speed measurement is pretty accurate (much
better than position);
3) heading can be accurate if ground speed is greater
than 0;
4) height estimate is by far the most 'problematic'
measurement estimation that such sensors can
do so be carefull to use this data for nothing more than
a reference (i.e. autopilot cannot use
this data for landing purposes 
);
);From: Haiyang Chao
Sent: Wednesday, September 09, 2009 5:08 AM
To: address@hidden
Subject: Re: [Paparazzi-devel] Re: Interfacing to the EagleTree
airspeed sensor
Hi, good work for the air speed sensor.
Ublox GPS does give an estimate regarding the GPS data accuracy. In "NAV_SOL" msg (GPS_SOL on pprz GCS), there is Pacc(3D Position Accuracy Estimate ) and Sacc(Speed Accuracy Estimate ). Of coz, we have to assume the manufacture estimate is good.
Haiyang
On Tue, Sep 8, 2009 at 8:27 PM, Vassilis V. <address@hidden> wrote:
Roman,I was thinking about this, but my concern is that I am not sure how much confidence I should have on the GPS data. The speed should be relatively accurate but it is ground speed so any wind gusts or changes in the attitude will introduce an error related to the airspeed. I never had too much confidence on the GPS reported altitude. I can upload the CSV files from my test flights if you would like to have a look at the raw data. Keep in mind that at this point I had to reverse engineer the I2C protocol and conversion of the raw values to altitude and airspeed so I can not be certain on the calculations done by the sensor itself.A few days ago I finally got a response from Bill Parry of EagleTree and he offered to give me some information. I will let you know when I get more details.A few emails from those of you interested may give Bill him an extra intensive to be more forthcoming ( billpa_at_eagletreesystems_dot_com)Regards,Vassilis
On Sep 8, 2009, at 4:40 PM, Roman Krashanitsa wrote:
Hi Vassilis,Good job on the airspeed sensor. Did you do any finer analysis on error estimates for the sensor? From the plot, I can see that your GPS measurements were not very stable and/or airspeed measurements are very inertial. Is it true?What was the mean and deviation of the difference between GPS and airspeed readings in a stable portion of the flight?Roman2009/9/7 Vassilis V. <address@hidden>
I got both the airspeed and altimeter sensors from EagleTree working. The source code and some data can be found at:
http://vrhome.net/vassilis
At this point I am simply passing the airspeed and altitude through the telemetry down to the ground station. The next step is to work on fw_v_ctl.c to allow the airspeed to be used for controlling the throttle (i.e. maintain constant airspeed). This will take me a bit longer, I will let you know when I have something working.
Vassilis
On Sep 6, 2009, at 9:59 AM, Vassilis V. wrote:
All,
I had a successful flight with the EagleTree airspeed module ( http://www.eagletreesystems.com/Support/manuals/airspeed-v3.pdf ) connected to my TWOG. I did not have to do any hardware modifications, the sensor has an I2C interface that I connected directly to the TWOG I2C port. I was able to figure out the slave address and decode the raw output of the sensor. The results are quite good, the module is low cost ($43) and comes with a very good pitot tube that includes static and dynamic ports. For now I am simply sending the airspeed through the telemetry, the next step would be to modify the autopilot code to regulate the throttle in order to keep the airspeed constant (and a minimum ground speed). I will do the same with the altimeter sensor.
See http://vrhome.net/vassilis for more information, I will publish the code as soon as I have the altimeter tested (probably in a week or so). Is there any interest on getting this released code released into CVS?
Vassilis
_______________________________________________
Paparazzi-devel mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/paparazzi-devel
_______________________________________________
Paparazzi-devel mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/paparazzi-devel
_______________________________________________
Paparazzi-devel mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/paparazzi-devel
_______________________________________________
Paparazzi-devel mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/paparazzi-devel