I've put my project on hold due to other priorities but the last I worked on it was with a MPU-9150 board and code supplied by Pansenti. This worked very well in my quick tests.
just one detail: do you think per your experience it could be useful to pair this board with a pressure sensor like https://www.sparkfun.com/products/11084 to give a complete data set to the sw libraries I've seen around for 10 DOF boards ?
I approached recently the subject and don't know yet if this additional data is actually needed to obtain additional precision, just in case... :-)
Al Testani > Mario CannistràJune 30, 2014 at 9:03am
I have only used barometric sensors in weather station applications so really don't know. It depends on whether the pressure is actually fused to the other data by the software.
What I was trying to do is to enhance the heading precision and further reduce drift by fusing in GPS heading data. I never got to that point because of the lack of fusion software to do that and the project went on hold. My application requires precise and fast heading data.
that pressure sensor gives direct compensated reading of altitude in 20-bit measurement (meters): i will check if the recent sw libraries use that.
i'm also interested in fast and accurate heading data since i will try to log position data and correct from to time to time with gps only if i see it's actually needed. i've found different opinions on the feasibility so ... will try and see.
will update this thread once I'll have first results.
I see that your message is 1 year old, but I'm currently searching for a good IMU solution (hw+sw) and read many of your posts, until I came to this one. It seems that OpenIMU did not go in production: did you find an alternative solution or did you solve the drift issue with ArduIMU sw/hw ?
It's now more than 1 week that I'm looking at various boards and would really like to find one for my specific project: I need to log (on a microSD) all gyro, accel, compass and barometer data for later processing. The logging sensors will be placed on a human subject moving freely (running, jumping, going on a rollerblade, etc.). I would also like to log also GPS data for later processing, possibly allowing IMU+GPS data processing to fix intertial errors with GPS data every N seconds.
I understand that some cumulative error could always be present, but I would love to avoid wasting money on devices that do not even allow the software processing with usual algorithms because of excessive continuous drift at the source (if I'm correctly interpreting some forum and blog posts around the web).
" I am very saddened to tell you that Fabio Varesano, the manufacturer of FreeImu died on Christmas Day. He was only 27 years of age. This morning we attended his funeral. I remember him as a friend with a great desire to try and help others. "
....
" No he was not sick, a cerebral infarction. "
I am speakless - my deep condolences to the family.
Sorry to be asking simple questions, but do we still use the Arduino IDE with this code? I got lost with the discussion about Python since I don't have/use that. Where should we download a version that should work for ArduIMUv3 and the Arduino 1.1 IDE. Thanks for the help!
Replies
Mario,
I've put my project on hold due to other priorities but the last I worked on it was with a MPU-9150 board and code supplied by Pansenti. This worked very well in my quick tests.
Pansenti has taken down all of their stuff (not sure why) but I found this library on GitHub:
https://github.com/richards-tech/MPU9150Lib
There are several MPU9150 breakout boards available.
Hope this helps.
Thank you Al, really appreciated.
I immediately searched and found an MPU 9150 board at sparkfun https://www.sparkfun.com/products/11486
just one detail: do you think per your experience it could be useful to pair this board with a pressure sensor like https://www.sparkfun.com/products/11084 to give a complete data set to the sw libraries I've seen around for 10 DOF boards ?
I approached recently the subject and don't know yet if this additional data is actually needed to obtain additional precision, just in case... :-)
I have only used barometric sensors in weather station applications so really don't know. It depends on whether the pressure is actually fused to the other data by the software.
What I was trying to do is to enhance the heading precision and further reduce drift by fusing in GPS heading data. I never got to that point because of the lack of fusion software to do that and the project went on hold. My application requires precise and fast heading data.
that pressure sensor gives direct compensated reading of altitude in 20-bit measurement (meters): i will check if the recent sw libraries use that.
i'm also interested in fast and accurate heading data since i will try to log position data and correct from to time to time with gps only if i see it's actually needed. i've found different opinions on the feasibility so ... will try and see.
will update this thread once I'll have first results.
Cheers,
Mario
Here is a link that my help you. You can use the V3, GPS & Barometric Sensor to get what you need.
DCM V3 code
http://diydrones.com/forum/topics/arduimu-v3-dcm-code-with-calibration
Thank you Al.
Will have a look at the code asap.
Cheers,
mario
Hi Al.
I see that your message is 1 year old, but I'm currently searching for a good IMU solution (hw+sw) and read many of your posts, until I came to this one. It seems that OpenIMU did not go in production: did you find an alternative solution or did you solve the drift issue with ArduIMU sw/hw ?
It's now more than 1 week that I'm looking at various boards and would really like to find one for my specific project: I need to log (on a microSD) all gyro, accel, compass and barometer data for later processing. The logging sensors will be placed on a human subject moving freely (running, jumping, going on a rollerblade, etc.).
I would also like to log also GPS data for later processing, possibly allowing IMU+GPS data processing to fix intertial errors with GPS data every N seconds.
I understand that some cumulative error could always be present, but I would love to avoid wasting money on devices that do not even allow the software processing with usual algorithms because of excessive continuous drift at the source (if I'm correctly interpreting some forum and blog posts around the web).
Mario
This is very sad what i found here:
http://www.multiwii.com/forum/viewtopic.php?f=8&t=649&start...
" I am very saddened to tell you that Fabio Varesano, the manufacturer of FreeImu died on Christmas Day.
He was only 27 years of age.
This morning we attended his funeral.
I remember him as a friend with a great desire to try and help others. "
....
" No he was not sick, a cerebral infarction. "
I am speakless - my deep condolences to the family.
Rob
Am I supposed to use that calibration.h you supplied and/or am I supposed to run the calibration application?
Thanks for the gyro fix.
Dave
Sorry to be asking simple questions, but do we still use the Arduino IDE with this code? I got lost with the discussion about Python since I don't have/use that. Where should we download a version that should work for ArduIMUv3 and the Arduino 1.1 IDE. Thanks for the help!
Dave