Abstract Angular accelerometer models estimated from the frequency response data using the Box–Jenkins method, where the calibration measurement system disturbance is parametrised independently from the dynamics. The models are identified from the frequency response measurement series using two types of angular acceleration input reference. Delays between these two references with the sensor data are approximated using the cross-correlation. The measurement data treated as a multi-channels series, which resulted in the concatenated model. The results demonstrate that the model with angular acceleration estimate as the input yields a higher degree model compared to the model with angular acceleration command as the input. The simpler model structure is possible by considering lower order models at the comparative Akaike Information Criterion value range.
Characterising Angular Accelerometer Calibration Setup Disturbance Using Box–Jenkins Method
2017-12-15
16 pages
Article/Chapter (Book)
Electronic Resource
English
Accelerometer Bias Calibration Using Attitude and Angular Velocity Information
Online Contents | 2016
|British Library Conference Proceedings | 1997
|Accelerometer Bias Calibration Using Attitude and Angular Velocity Information (AIAA 2016-1140)
British Library Conference Proceedings | 2016
|