Accelerometer Bias Self-Calibration: Laboratory Demonstration of a MEMS Accelerometer with Part-Per-Million Bias Error

Michael Bulatowicz

Abstract:	Accelerometer bias self-calibration is a potential game-changer for high-performance low-cost accelerometers. This paper describes the concepts, methodology, and demonstration of technology for self-calibration of the actuation-induced component of accelerometer bias. This method and the associated algorithms are independent of the design of the sensitive element, and place the burden of self-calibration on the implementation in the electronics. As a result, this method and algorithm are applicable to a wide variety of accelerometer designs already on the market. Preliminary results have demonstrated residual bias errors (after application of a fourth-order thermal model) at approximately 0.6 parts per million RMS of full-scale range using a sub-optimal mechanization and an accelerometer that had been rejected from the production line for excessive bias error.
Published in:	Proceedings of the ION 2015 Pacific PNT Meeting April 20 - 23, 2015 Marriott Waikiki Beach Resort & Spa Honolulu, Hawaii
Pages:	1020 - 1023
Cite this article:	Bulatowicz, Michael, "Accelerometer Bias Self-Calibration: Laboratory Demonstration of a MEMS Accelerometer with Part-Per-Million Bias Error," Proceedings of the ION 2015 Pacific PNT Meeting, Honolulu, Hawaii, April 2015, pp. 1020-1023.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In