search

ozzmaker / BerryIMU

Code related to BerryIMU
http://ozzmaker.com
184 stars 108 forks source link

used this with MPU9250, modified accordingly and works but issue with heading #17

Open bbelmar opened 4 years ago

bbelmar commented 4 years ago

Hello there, first of all thank you so much for this amazing work! you are a life saver.

I used this and modified to get the readings from MPU9250. Accel works as expected, I was expecting the gyro to read similar magnitudes as the accel but the magnitudes are way smaller, even if I tilt the sensor -90Deg in the Y axis , the gyro measures only approx -6., but the accel measures approx -90. I thought the code was turning gyro information in Degrees? The accel in X goes crazy but i dont know if that is an expected reading.

Accel in X= 146.01047352241366 Accel in Y= -94.48802755114491 ##this is when i rotated the sensor to the left 90 deg approx GyroX = 0.0958970146179199 GyroY = -6.615839959945679 ## shouldn't this be same reading as accel in Y? GyroZ = 0.010706434631347735 Heading = 0.4136347527893839 Compensated Heading = -496.89575038831293

And lastly, as you can see the heading is not responding well... If i rotate the sensor 180 deg, it goes from approx 25-> 0 -> 40 as i rotate...

I know im asking a lot, but if you have the time to reply would be amazing. Thank you so much!

bkfichter commented 3 years ago

Did you ever get this working? For the BerryGPS-IMU v4, I have been using the python versions of the calibration and "compass" and the headings are nothing usable. And tilt compensation doesn't do anything usable either.

peterp-ozzmaker commented 3 years ago

Did you ever get this working? For the BerryGPS-IMU v4, I have been using the python versions of the calibration and "compass" and the headings are nothing usable. And tilt compensation doesn't do anything usable either.

Have you tried to graph out your results after calibration to confirm that the calibration is correct? Calibration guide

bkfichter commented 3 years ago

Well, I did run calibrateBerryIMU.py (multiple times) and took the results and applied to berryIMU.py. I just checked to make sure the code was implementing the calibration results. It appears to be. I haven't actually graphed the results. I will do this today. I'm not sure what this will give me if I am already running the calibration script and applying the results. Will be interesting to see the graph though.