Developer

MPU6000 scaling issue

 

     Thanks to help from Sami Finnila and Wesley Chuen we've found an issue with the default MPU6000 accelerometer scaling in some of the recent APM2 boards shipped in the last month or so.  You can find the original discussion from Sami here.

     Internally we use "DCM" to merge the gyro and accelerometer values together to calculate attitude.  With the recent boards we are finding the accelerometer values are 2x as big as they should be which leads to an inaccurate attitude estimate.  It's apparently possible to fly with this misconfiguration but it at least makes "the leans" more common (you can recognise "the leans" by your copter requiring more and more manual roll or pitch to keep it level as you fly).

 

UPDATE (11-May-2012):

     Craig and Tridge have completed a patch to automatically detect the scaling of the accelerometer so we can compensate.

     This patched version is being called 2.5.5-Beta and is in the downloads area.  After a short testing period (maybe 24h ~ 48h), if all goes well it will be available in the mission planner.

     We're still working on the next official release 2.6 so we haven't included any improvements to flight dynamics in this patch except Adam's "Retro" loiter.  This uses the ground speed directly from the GPS instead of calculating it using lat/lon position.  You can enable this by setting the "RETRO_LOITER" parameter to "1".  BTW, it's "retro" because this is the method used in the popular 2.0.49 version of Arducopter.

 

     Feel free to post comments / issues regarding this patch release in the discussion below.

     If you have recieved an APM2 in the last month, please try the test shown above:

  • hold your APM2 flat and check the "az" value in the mission planner's Flight Data, Status screen is approximately 1000 (it should be for everyone, regardless of whether you have a new or old board)
  • hold your APM2 up-side-down (but still level) and check the "az" value again.  If it's not close to 1000 then we want your tlog!  :-)

 

     Next just roll and pitch your board a bit and attach your tlogs below.  These logs can be found in your AP Mission Planner's log directory.  On a windows machine you will likely find a recently updated file here: C:\Program Files (x86)\APM Planner\logs\YYYY-MM-DD HH-MM-SS.tlog

 

     Thanks for your help.

Tags: ArduCopter, ArduPlane, MPU6000, Scaling

Views: 5156

▶ Reply to This

Replies to This Discussion


DeveloperPermalink Reply by Randy on May 10, 2012 at 6:36am

Andke,

     Thanks for that.

     Yes, your bad gyro behaviour is absolutely related.  I've seen the same thing on my new APM2 board.  I believe the reason is that there's is I term build up on the PID controller that is used to combine the gyros values into the accelerometer.  Basically it's caused by the constant error that's appearing because the accelerometer values are so far from the gyro (and we're assuming the accelerometers are correct).

    

     Ok, thanks all for your logs.  I think we're good now.  Tridge and Craig have checked in a change to trunk today that we think will resolve the problem.  I've just tested it and it appears ok so next we will get some real testers to give it a go and then we will push out a patch release (2.5.5).

Permalink Reply by Tony Lu on May 9, 2012 at 8:16pm

Same here.

Attachments:
Permalink Reply by Edwin on May 10, 2012 at 12:55am

Here are some logs for boards affected by the issue.

Attachments:
Permalink Reply by Eagle on May 10, 2012 at 4:47pm

Now that's what I call community work will all tlogs replies!

I just did the test even if my APM2 comes from pre-ordered batch. ... just for my curiosity.  Looks good az=+-1000.

Thanks Randy and the Community.

Permalink Reply by Florin on May 10, 2012 at 6:59pm

I have the exact same issue, upside down the offset is 3000.

I'm using my APM 2.0 in a plane (arduplane 2.33) anyone can tell me if this defect will affect the dynamic of the flight?


DeveloperPermalink Reply by Randy on May 10, 2012 at 9:27pm

Florin,

     Yes it will affect the flight dynamics.  From the feedback I've seen, it's likely not enough to cause a crash but it will have an incorrect attitude information so I'm pretty sure it won't fly waypoints as well.

     The ArduPlane patch release will also be pushed soon.  Probably before the weekend is over.

Permalink Reply by Florin on May 11, 2012 at 5:49am

Thanks for the quick update Randy.

Permalink Reply by Mark Murray on May 11, 2012 at 1:57am

Please confirm that the APMv2/FlowSensor patch is also NOT in this 2.5.5 release?

Thanks

M

Permalink Reply by andycross1 on May 11, 2012 at 5:50am

Hello Mark

What was your reference to the Flow Sensor I though it was working or am I mistaken?

Permalink Reply by Mark Murray on May 11, 2012 at 5:59am

Andy,

The flow sensor has been relocated on the APMv2, and is currently advertised as "supported in 2.6+". I'm hoping that a patch release will provide this support earlier, if 2.6 is going to be delayed for any reason.

M

Permalink Reply by andycross1 on May 11, 2012 at 7:10am

Hi Mark

Thanks for your reply, do you know if they are changing where you connect to the APM2 as it's not the easiest place to solder connections to?

Permalink Reply by Mark Murray on May 11, 2012 at 7:15am

Andy

All I know is what I see here:

http://code.google.com/p/arducopter/wiki/AC2_OptFlow

:-)

M

RSS

Social Networking

Top Content 

1 APM 2.5 from hobbyking

APM 2.5 from hobbyking
2 Alt_hold and loiter acting crazy

Alt_hold and loiter acting crazy
3 Mapping Mission Ride Along

Mapping Mission Ride Along
4 Looking for best laptop screen to see in the sun

Looking for best laptop screen to see in the sun
5 APM Copter 2.9.1.b has been released

APM Copter 2.9.1.b has been released
6 ArduPlane 2.73 released

ArduPlane 2.73 released
7 New micro-cameras emulate insects' compound eyes

New micro-cameras emulate insects' compound eyes
8 Hobby King Ardupilot

Hobby King Ardupilot
9 3 axis Camera Gimble connections.

3 axis Camera Gimble connections.
10 Thirty Days (#17): 12-rotor close up

Thirty Days (#17): 12-rotor close up

Contests

Season Two of the Trust Time Trial (T3) Contest has now begun. The fourth round is an accuracy round for multicopters, which requires contestants to fly a cube. The deadline is April 14th.

A list of all T3 contests is here. 

Groups

Advertisement

© 2013   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service