Yaw Drift

[MrMikeGrant]

6DOF IMU's have well documented problems with yaw drift. The drift rate tends to be small (i.e. 0.25 degrees per min) but can add up over time (60 min = 15 degrees).

If accurate yaw angles are required, the only way around this issue may be to switch to a 9DOF instead.

[bndnio]

I believe the MABX actually has a gyro built into it. I assume this would have the measures in place to prevent Yaw drift, so we should definitely look into that as well. And the gyro board were using I believe has a magnetometer (compass) in it, so I believe it handles that issue itself.. That thing is really a black box though, documentation is hard to find for it.

For whoever picks this project up, you should:

• [ ] Check MABX documentation to see if it has it's own gyro/magnetometer and the gyro board is redundant
• [ ] If not, look into the documentation to see if yaw drift is handled on the board itself
• [ ] If not, check the existing firmware to see if it handles yaw drift correction on the Arduino
• [ ] If not, look for new/updated firmware (the existing code uses open source firmware which was not complete at the time of the project)

[MrMikeGrant]

Cool I did not know the MABX has onboard sensors!

1. Looked at the tech specs on website and it lists the MABX has "Motion sensing with 3-axis accelerometer and Pressure sensing for altitude indication"

2. THe MPU 6050 is a 6DOF IMU with 3-axis accelerometer and 3axis gyro, unfortunately it does not contain the magnetometer required for yaw stability. This is the same sensor we have on the green ship group data acquistion project and we ran into the same problem. We are currently looking at viable alternatives which are 9DOF IMU's with solid documentation and opensource code available.