Blogs

APMLogVisualizer version 1.32, with C# Source

Posted by Dan Gray on July 22, 2012 at 6:30pm

I've had quite a few positive comments, and also, some folks have pointed out bugs and made suggestions for improvements.
So, I fixed the bugs (all that I know of anyway), added most of the suggested features, and added a couple new features of my own.

Above is a YouTube Video that shows most of the features, and how I narrowed down the reason for my crash.

Here's the latest software:
http://siderealtechnology.com/APMLogVisualizer1.32.zip

In the following YouTube video, you can see the crash, as I had my GoPro on the copter:

E-mail me when people leave their comments –
Follow

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

Join diydrones

Comments

  • T3
    Krzysztof Bosak September 1, 2012 at 12:07am

    Hi all, I am browsing this:

    http://code.google.com/p/ardupilot-mega/wiki/MPDatalog

    and clearly the description is not explaining how

    ATT, 0, 62, 0, 295, 0, 2340, 2351 (which I see in examples)

    translates into

    Roll    Roll Attiude (phi)    deg
    Pitch    Pitch Attitude (theta)    deg
    Yaw    Yaw Attitude (psi)    deg

    this concerns all data lines.

    Also for some magic reason the log files have format CTUN, ATT, GPS,  etc

    while

    Log.pde contains code like

    Serial.printf_P(PSTR("GPS: %ld, %d, %d, %4.7

    in general all in the format XXX: and not XXX,

    Where can I found the ultimate guide saying which log is log and what log is telemetry then log when parsed as telemetric log?

    Why nobody uses examples in the docs, it would be much better than incomplete specs of the format, at the moment one cannot guess which document is talking about what you need...

  • Moderator
    Ruwan July 31, 2012 at 1:07pm

    Hi, Dan

    It'll be awesome if you can share it in http://code.google.com/ when you share the code. Again, thanks for the great effort.

    regards,

    Ruwan.

    Google Code
  • Moderator
    Graham Dyer July 31, 2012 at 12:58pm

    Thanks for the reply and explanations, it's great to be able to visualize the flight like this and thanks for fixing those little things. Maybe this will be incorporated in the Mission Planner one day.

  • Dan Gray July 31, 2012 at 11:29am

    Hi Graham,

    I've thought of all those features (they're all really great ideas), but I don't have time to add them (except for the Bad Lines, which I've already fixed in version 1.34).

    Oh, btw, that wasn't actually a bug.  In Arducopter version 2.7, the format changed for the CTUN line, the new version stopped logging the value ManualBoost, for some reason, and this made APMLogVisualizer reject it as a bad line.

    I think I fixed your last bug too, after loading a file again.

    I wish channels were a part of the LOG, but the only channel I see is Throttle_In, but no other inputs.  The 4 Motors are the outputs for channel 1 to 4 I think, but you have to turn on the various log options.

    Also, while we're on the subject, I would add the PID values, it seems to be a logging option, but there are no PID values in the log files, even when I have PID enabled.

    Anyway, I'll send the new version out later today, and I've decided to post the code publicly, so anybody can download and work on the software.


    Dan Gray

  • Moderator
    Graham Dyer July 31, 2012 at 1:00am

    Hi Dan.

    Thanks for this great piece of software.

    I'm getting the following: "Your Log File has about 9408 Bad Line(s) in it.", what causes this?

    A few suggestions/requests:

    • Full screen mode to view on a large monitor, increasing size of track area
    • Zoom in to any portion of the track,
    • Possibility to log Channel 1-7 input and outputs (although that may be the bad lines that I'm missing)
    • 3D view of quad as well as separate roll & pitch
    • If I load a second log or the same one again there's no display of the track

  • Dan Gray July 26, 2012 at 2:08pm

    Hi Hein,

    It's a home brew power distribution / Isolated voltage, and Isolated current circuit board.

    Here's pictures:

    http://siderealtechnology.com/VoltCurrentBoardByItself.jpg

    http://siderealtechnology.com/VoltCurrentBoardInstalled.jpg

    Here's eagle files:

    http://siderealtechnology.com/QuadPwrDist.zip

    Dan

  • 100KM
    Hein du Plessis July 25, 2012 at 12:23pm

    Thanks Dan, though - what do you use to measure the Amps? Or is it built into the APM2?

  • Ozan Aktaş July 25, 2012 at 3:06am

    Nice work! Yes it will be great if this tool is integrated into MP.

  • Dan Gray July 24, 2012 at 5:49am

    Hein, The parameter for logging amps is Current, but to enable it, you have to say

    "Enable Cur", or else Enable Curr, I can't remember 1 or 2 R's, and it's not documented properly, I had to figure it out myself.

    Ruwan, Yes, it could possibly be integrated into Mission Planner, I'm ok with it.

    Angel, Hmmm, the yaw representation moves on mine, can you send me a log file where the yaw doesn't move?

    Dan

  • afernan July 24, 2012 at 12:08am

    @Dan:

    thank you for this fantastic analysis tool. This is really wellcome for developers and accelerates a lot the work.

    For your information, I´ve got this message while loading an APM2 log file (from AC 2.6.1)

    Your Log File has about 13750 Bad Line(s) in it.
    CTUN,0,0,-10,0,0,0,-9,0,0,0
    CTUN,0,0,-9,0,0,0,0,0,0,0
    And also, the yaw representation doesn´t move (although yaw data is in).

    Anyway all the rest works perfectly. A nice addition should be to add other logs type, like PID.

    Best regards

    Angel

This reply was deleted.