AN ANFIS-Based Modeling of Thermal Drift of MEMS-Based Inertial Sensors

W. Abdel-Hamid

Abstract:	The last two decades have shown an immerging trend of using micro-electro-mechanical sensors (MEMS) in many navigation applications particularly personal and land vehicle navigation. The low cost, lightweight and miniature size of MEMS inertial sensors are basic requirements that attract many manufacturers in Automobile market. However, due to their small size and other specific fabrication process, there are some limitations on the performance of MEMS-based inertial sensors which will consequently jeopardize the overall accuracy of navigation systems implementing them. One of the most major shortcomings of MEMS sensors is the significant change of their performance characteristics (e.g. bias and scale factor) with environmental factors variation especially change in temperature. This necessitate the development of an accurate and reliable thermal model, describing sensor drift due to thermal variation, to be used for either on-line or post processing applications. In this paper, the results of lab thermal testing of two MEMS-based IMUs (Cross-bow AHRS400Ca and MotionPak II) are presented first. Then, the Adaptive Neuro-Fuzzy Inference System ANFIS is adopted to develop a new model that appropriately describes the thermal variation of each sensor output.
Published in:	Proceedings of the 17th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2004) September 21 - 24, 2004 Long Beach Convention Center Long Beach, CA
Pages:	784 - 791
Cite this article:	Abdel-Hamid, W., "AN ANFIS-Based Modeling of Thermal Drift of MEMS-Based Inertial Sensors," Proceedings of the 17th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2004), Long Beach, CA, September 2004, pp. 784-791.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In