Quote:
Originally Posted by Richard100
Austin - Did you conduct the user calibration process or simply accept the factory calibration values?
Did you make use of the magnetometer?
|
We did a startup calibration (save the last 6 seconds of data, and re-zero when the robot enables), and did not use the magnetometer. Exploring factory calibration accuracy sounds like a good idea, thanks! We didn't really have time to explore many options during the season and opted to do what we knew from experience would just work.
Quote:
Originally Posted by Richard100
The ADIS16480 has embedded sensor fusion, running an on-board Extended Kalman Filter. This puts it functionally on par with the Bosch BNO055. These 'Orientation Sensors', with embedded fusion algos, have ready low (or no) drift orientation outputs as well as stabilized gravity vector outputs - making for simpler/faster integration. (The ADI device appears to retail > $2k.)
|
I get to write kalman filters at my day job on good days
For us, I'm less worried about embedding the sensor fusion algorithms in the sensor as I am about the quality of the sensor. A pre-built EKF will work well in the cases that it was designed for. By knowing your application more specifically (like that you have wheels and tend to move forwards, etc), you can make different tradeoffs in your filter design. A device with a pre-built EKF does save a bunch of time though when it works, as you aptly point out.
Juan recommended the ADIS164
60 as an accurate sensor to be used by an EKF on the roboRIO.