Scholarlink

To reduce Temperature Drift Errors (TDE) and enhance environmental adaptability of Capacitive MEMS Gyros (CMG) in complex conditions, a modified TDE estimation system is presented. Firstly, linear structural deformation of Si-based material is studied to reveal TDE’s root cause and explore all-new complete Temperature Correlated Quantities (TCQ), temperature variation ΔT as well as its square ΔT2. Secondly, thermal experiment is designed to obtain more accurate TCQ and TDE, and an optimized TDE precise estimation model is built using Radical Basis Function Neural Network (RBFNN) to represent their nonlinearity. Thirdly, overall hardware design of the modified system is implemented. Finally, the modified system is tested in thermal experiments and its optimized model is compared by the conventional models based on ΔT and least square method or RBFNN in bias stability. The experimental results show CMG is improved by 10% in bias stability, which means the modified system estimates more precise TDE to enhance its environmental adaptability markedly in complex conditions.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Implementation of a Modified Temperature Drift Errors Estimation System for Capacitive MEMS Gyros

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Qi Bing,
  • Li Peng,
  • Guan Menghan

摘要

To reduce Temperature Drift Errors (TDE) and enhance environmental adaptability of Capacitive MEMS Gyros (CMG) in complex conditions, a modified TDE estimation system is presented. Firstly, linear structural deformation of Si-based material is studied to reveal TDE’s root cause and explore all-new complete Temperature Correlated Quantities (TCQ), temperature variation ΔT as well as its square ΔT2. Secondly, thermal experiment is designed to obtain more accurate TCQ and TDE, and an optimized TDE precise estimation model is built using Radical Basis Function Neural Network (RBFNN) to represent their nonlinearity. Thirdly, overall hardware design of the modified system is implemented. Finally, the modified system is tested in thermal experiments and its optimized model is compared by the conventional models based on ΔT and least square method or RBFNN in bias stability. The experimental results show CMG is improved by 10% in bias stability, which means the modified system estimates more precise TDE to enhance its environmental adaptability markedly in complex conditions.