|
Re: Got Gyro drift? Try Sensor Fusion !
Steven - You need to worry about both. I actually developed the code with the sensor installed on our last year's robot (I do this to ensure I'm accounting for all the other code we run, as the rate gyro numerical integration is particularly time sensitive). In limited testing, I found no real issue installing/operating the IMU on my robot until the drive system was engaged. I also found that pushing (an off) robot alongside my IMU-capable robot produced a similar field disturbance (which makes an interesting robot proximity detector).
Of course you would want to locate the IMU as far away as possible from both the magnets in the DC motors and the current supply wiring - 33 cm is too close. I also experimented briefly with shielding the CIMS with mu-metal, with some encouraging but admittedly incomplete results.
As I mentioned, 'challenges remain in successfully integrating the Magnetometer on an FRC bot', and 'be sure you actuate all motors and mechanisms to test for interaction with the Yaw indication, which uses the Mag'. Hint: You don't have to use the Mag all the time ...
|