Copter-3.3 beta testing

Warning #1: an issue has been found with Tower's Pause button which can cause the vehicle to fly to an old position if the vehicle has not sent a position update to Tower in some time.

Warning #2: Copter-3.3.2 fixes a bug found in Copter-3.3.1's desired climb rate initialisation which could lead to a sudden momentary drop when switching from Stabilize or Acro to AltHold, Loiter or PosHold.

Warning #3: Copter-3.3.2 fixes an issue found in Copter-3.3.1 which could lead to hard landings in RTL or AUTO if the WPNAV_SPEED_DN was set too high (i.e. >400 or 4m/s) and/or the WPNAV_ACCEL_Z was set too low (i.e. <100 or 1m/s/s).

Warning #4: a bug was found in Copter-3.3 which could cause a sudden crash if you abort a Take-off initiated from a ground station. Video description is here. The bug is fixed in Copter-3.3.1 so we recommend upgrading.

Note #1: AC3.3-rc8 corrected a long standing bug in the HDOP reporting. HDOP values will appear about 40% lower than previously but this does not actually mean the GPS position is better than before.
Note #2: if upgrading from AC3.2.1 the vehicle's accelerometer calibration needs to be done again.
Note #3: set SERIAL2_PROTOCOL to "3" and reboot the board to enable FrSky telemetry like in previous versions.
Note #4: the wiki will be updated over the next few weeks to explain how to use the new features

Copter-3.3.1 is available through the mission planner. The full list of changes vs AC3.2.1 can be see in the ReleaseNotes and below are the most recent changes since AC3.3.

Sadly this version (and all future versions) will not run on the APM2.x boards due to CPU speed, flash and RAM restrictions.

Changes from 3.3:

1) Bug fix to prevent potential crash if Follow-Me is used after an aborted takeoff

2) compiler upgraded to 4.9.3 (runs slightly faster than 4.7.2 which was used previously)

Changes from 3.3-rc11:

1) EKF recovers from pre-arm "Compass variance" failure if compasses are consistent

Changes from 3.3-rc10:

1) PreArm "Need 3D Fix" message replaced with detailed reason from EKF

Changes from 3.3-rc9
1) EKF improvements:
    a) simpler optical flow takeoff check
2) Bug Fixes/Minor enhancements:
    a) fix INS3_USE parameter eeprom location
    b) fix SToRM32 serial protocol driver to work with recent versions
    c) increase motor pwm->thrust conversion (aka MOT_THST_EXPO) to 0.65 (was 0.50)
    d) Firmware version sent to GCS in AUTOPILOT_VERSION message
3) Safety:
    a) pre-arm check of compass variance if arming in Loiter, PosHold, Guided
    b) always check GPS before arming in Loiter (previously could be disabled if ARMING_CHECK=0)
    c) sanity check locations received from GCS for follow-me, do-set-home, do-set-ROI
    d) fix optical flow failsafe (was not always triggering LAND when optical flow failed)
    e) failsafe RTL vs LAND decision based on hardcoded 5m from home check (previously used WPNAV_RADIUS parameter)

Thanks for your testing!

Maybe the cf is the opposite of what we need...

According to Leonhard's explanation:

The formulae for standard deviation is sqrt(sum(samples^2)/No_Samples) In our case we don't want to include normal motion of the copter so

samples = HighPassFilter(IMU).

We also want an output that represents the recent performance of the copter, not the whole flight. So we replace the sum()/No_Samples with a low pass filter that effectively looks at a decaying average over the last N samples.

So this becomes a moving standard deviation over the last segment of time where that time is defined by the time constant of the low pass filter.

maybe we should calculate something like:

samples * sqrt(sum(samples^2)/No_Samples)

This would account for the actual (high-pass filtered) level of vibration and its variation.

In probability theory and statistics, the coefficient of variation (CV), also known as relative standard deviation (RSD), is a standardized measure o…

Replies

- - - T3
      
      Thorsten > Rob_Lefebvre July 2, 2015 at 10:12am
      
      Maybe the cf is the opposite of what we need...
      
      According to Leonhard's explanation:
      
      The formulae for standard deviation is sqrt(sum(samples^2)/No_Samples) In our case we don't want to include normal motion of the copter so
      
      samples = HighPassFilter(IMU).
      
      We also want an output that represents the recent performance of the copter, not the whole flight. So we replace the sum()/No_Samples with a low pass filter that effectively looks at a decaying average over the last N samples.
      
      So this becomes a moving standard deviation over the last segment of time where that time is defined by the time constant of the low pass filter.
      
      maybe we should calculate something like:
      
      samples * sqrt(sum(samples^2)/No_Samples)
      
      This would account for the actual (high-pass filtered) level of vibration and its variation.
    - T3
      
      Thorsten > Rob_Lefebvre July 2, 2015 at 10:02am
      
      I guess we need some mixture.Maybe the coefficient of variation to make it comparable between different setups.
      
      But somehow it seems we need to analyze more logs...
      
      Coefficient of variation
      
      In probability theory and statistics, the coefficient of variation (CV), also known as relative standard deviation (RSD), is a standardized measure o…
      - Rob_Lefebvre > Thorsten July 2, 2015 at 10:11am
        
        Yes, definitely need too look at more data. This is largely why I posted this, to get the discussion going. Maybe needs it's own thread or something.
        
        Leon DeBell > Rob_Lefebvre July 2, 2015 at 11:20am
        
        If it's of any use; my quad has pretty low vibes...circa 0.5-1 and I'm still on RC5...I can flash RC7 tomorrow morning and do a quick loiter and upload both an RC5 and RC7 log?
        
        T3
        
        Thorsten > Rob_Lefebvre July 2, 2015 at 10:13am
        
        Yes! Both: important and interesting.
  - T3
    
    Thorsten > Thorsten July 2, 2015 at 9:15am
    
    Maybe the difference comes from different logging rates. I was flying rc6.
    - Rob_Lefebvre > Thorsten July 2, 2015 at 9:30am
      
      Both tests were done on RC6.
- Rob_Lefebvre > Rob_Lefebvre July 2, 2015 at 8:58am
  
  And then here's the really interesting thing. Here's a really zoomed in graph of the electric vs gas accel data. Electric on top, gas on the bottom. The electric is showing clear sine waves, I believe you can actually see the rotor going around. You can actually see a Sine and Cosine wave on X and Y, always exactly 90 degrees out of phase as you'd expect, which is what tells me this it the rotor going around.
  
  The gasser however, looks like random noise. There's not much signal here. I mean, it almost looks like breaks in the data, (it doesn't have a fast card) but I'm not sure. My understanding is that shows up as bigger gaps in the data, which is not what we have here. So assuming that's not the problem, it's clear this data quality is much lower than the electric. But the electric shows much higher Vibe RMS data.
  - Doug Walmsley > Rob_Lefebvre July 2, 2015 at 9:09am
    
    Can you swap the memory card over to the gasser and rerun the test or at least put in the exact same memory card as in the electric version? I see similarities as report yesterday on my copter. Curious to see the results.
    - Rob_Lefebvre > Doug Walmsley July 2, 2015 at 9:31am
      
      Doug, was your missing data on the same sort of scale? I understood the missing data was more like, 1 second every few seconds. This looks more like, a few milliseconds every few milliseconds.
      
      But yes, I should try new cards.

of 558

This reply was deleted.

Discussions

Copter-3.3 beta testing

Replies

Activity