Dr. Stephen Bruder
The VectorNav VN-200 is a MEMS navigation system that includes an accelerometer and gyroscope, as well as other navigation sensors. The accelerometer and gyroscope alone make up a commercial grade inertial measurement unit (IMU). Commercial grade IMUs produce poor performance on their own, and when used in navigation systems they can accumulate significant errors in less than a minute of operation. However, large increases in IMU accuracy can be achieved by properly calibrating out some of the IMUs deterministic errors. In this paper, methods are discussed for removing fixed biases, scale factor errors, and misalignment errors in both accelerometers and gyroscopes using a rate table. After discussing the methods for calibration, data collected from a VN-200 is used to calibrate the IMU. With the VN-200 calibrated, two separate tests using data collected from the sensor are performed to measure the increase in performance of the IMU as a dead-reckoning navigation system. In the two tests, the calibration reduced the calculated position error by 94.4% and 92.7% and the calculated attitude error by 96.6% and 94.7%. This substantial increase in accuracy by performing a simple calibration is so beneficial, it should performed for all accelerometers and gyroscopes whenever possible. At the very least, sensors should have their fixed biases removed before being used in navigation.
Scholarly Commons Citation
Ferguson, J. (2015). Calibration of Deterministic IMU Errors. , (). Retrieved from https://commons.erau.edu/pr-honors-coe/2