Blogs

Developer

Arducopter Flight data needed

Posted by Ryan Beall on January 26, 2011 at 5:02pm

3689386909?profile=originalI have been working on an improved altitude estimator for a while here but have been having issues with implementation.

 

The theory works fine but as seen above in the FFT of the accelerometer data, there is a significant ammount of vibration/noise in the sub 50Hz range.  This is due to out of balance props etc.  Needless to say, this proves to be very detrimental to estimating altitude and velocity changes when these vibrations are an order of magnitude greater than the actual accelerations felt by climbing and descending.  You cant really low pass filter or band pass filter this because you need everything below 60 Hz or so. 

 

So here is what I need from you guys:

Flight data that includes the following:

  • dt in ms between samples
  • phi,theta,psi
  • barometric altitude
  • sonar altitude (if available)
  • ax,ay,az

make sure you notate units etc.  The more information on the data the better.  Thanks for the help and I will post the results.  I'm really hoping to get some data from guys with slick clean vibe free quads. 

 

Thanks

Beall

 

E-mail me when people leave their comments –
Follow

You need to be a member of diydrones to add comments!

Join diydrones

Comments

  • robert bouwens February 2, 2011 at 11:02am

    back on track - the mini flies.

    but - way too heavy: 384 gramm.

    now can i look how to get the flightdata after the neccessary diet...

  • robert bouwens January 31, 2011 at 12:10pm

    aua - i just fried my one of motor outputs ...

    need to modifiy the code that channel 6 acts like channel 1.

    revision 1.4 of the mega board is required when using turnigy plush esc's!

    maybe - maybe ... i get it working

  • robert bouwens January 30, 2011 at 9:13pm

    one esc died - i have to wait until i have the replacement in.

    then i can collect some data - if it flies :)

     

  • Developer
    Randy January 30, 2011 at 3:58pm

    At the beginning of this thread I mention how to capture some data with ArduCopterNG..there were some mistakes in that.

     

    For #1 - this line is fine if you're using the version of ArduCopterNG from subversion (not from download area)

      Log_Write_Sensor(AN[0], AN[1], AN[2], AN[3], AN[4], AN[5], press_baro_altitude);

    but if you're using the download version you need to use this line:

      Log_Write_Sensor(AN[0], AN[1], AN[2], AN[3], AN[4], AN[5], press_altitude);

     

    Also for #2, this line:

      Log_Write_Attitude(roll,pitch,yaw);

    should be changed to:

      Log_Write_Attitude(ToDeg(roll),ToDeg(pitch),ToDeg(yaw));

     
  • Developer
    Ryan Beall January 29, 2011 at 8:45pm
    Awesome work. Can you send me the matlab for that!
  • T3
    William Premerlani January 29, 2011 at 2:08pm

    All,

    Keep in mind that, even if there is no noise or vibration and even if you have an ideal anti-aliasing filter, there is still a good reason to oversample: it improves the effective resolution of the A/D.

    Best regards,

    Bill

  • Bill Nesbitt January 29, 2011 at 8:31am
    Kári: I think the above video was using an ADXL325. However, the ADXL335 is effectively the same part but with a less stiff spring mass and I use them interchangeably. Both have a 5.5KHz resonance frequency. I'm currently doing analog filtering of the signal down to 20Hz using a 2nd order butterworth filter constructed with a spare OPAMP channel. Although this is not absolutely necessary, what you will find is that adding higher capacitance to the already fairly high impedance output of the ADXL causes the signal settle times to grow requiring a longer sample time. The longer the sample time, the less samples you can take in a given time period. This is compounded by the large number of sensors (12 in my case) that need to be serviced by limited ADC resources. I believe that the requirement for conversion of good clean, high resolution, accurate sensor signals is only achieved by extremely high sample rates with conservative sample times.

    geir: I would be happy to share details of my work, but this thread should not be hijacked. PM me, or if there is other interest, I could start another thread.
  • T3
    William Premerlani January 28, 2011 at 9:51am

    Ryan,

    Here is a link to a posting that you might find interesting, reporting the results of tests that Peter Hollands conducted at a sampling rate of 32,000 samples per second.

    Besides rejecting vibration, the higher sampling rate has also dramatically improved the accuracy of attitude estimation and acceleration measurements which are key factors in dead reckoning computations.

    Best regards,

    Bill

  • Bill Nesbitt January 28, 2011 at 9:00am

    John:


    No, it is a design of my own built around an ARM Cortex M3 uC.

  • John Reynolds January 27, 2011 at 7:50pm

    Bill:

     

    That quad video was awesome!  Loved the way that sucker just launches into those waypoints! 

     

    Just curious- are you using the same hardware (APM and shield) that Arducopter is using and just tweaked the software, or are you using something else?

     

    John

     

     

     

This reply was deleted.