High-accuracy Motion Estimation for MEMS Devices with Capacitive Sensors

Motion estimation;Capacitive sensors;MEMS devices;Kalman filter;Mechanical Engineering;Engineering;Mechanical Engineering

Chen, YiOkwudire, Chinedum Emmanuel ;

University of Michigan

关键词: Motion estimation;    Capacitive sensors;    MEMS devices;    Kalman filter;    Mechanical Engineering;    Engineering;    Mechanical Engineering;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/147568/davidsky_1.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

With the development of micro-electro-mechanical system (MEMS) technologies, emerging MEMS applications such as in-situ MEMS IMU calibration, medical imaging via endomicroscopy, and feedback control for nano-positioning and laser scanning impose needs for especially accurate measurements of motion using on-chip sensors. Due to their advantages of simple fabrication and integration within system level architectures, capacitive sensors are a primary choice for motion tracking in those applications. However, challenges arise as often the capacitive sensing scheme in those applications is unconventional due to the nature of the application and/or the design and fabrication restrictions imposed, and MEMS sensors are traditionally susceptible to accuracy errors, as from nonlinear sensor behavior, gain and bias drift, feedthrough disturbances, etc. Those challenges prevent traditional sensing and estimation techniques from fulfilling the accuracy requirements of the candidate applications.The goal of this dissertation is to provide a framework for such MEMS devices to achieve high-accuracy motion estimation, and specifically to focus on innovative sensing and estimation techniques that leverage unconventional capacitive sensing schemes to improve estimation accuracy. Several research studies with this specific aim have been conducted, and the methodologies, results and findings are presented in the context of three applications. The general procedure of the study includes proposing and devising the capacitive sensing scheme, deriving a sensor model based on first principles of capacitor configuration and sensing circuit, analyzing the sensor’s characteristics in simulation with tuning of key parameters, conducting experimental investigations by constructing testbeds and identifying actuation and sensing models, formulating estimation schemes is to include identified actuation dynamics and sensor models, and validating the estimation schemes and evaluating their performance against ground truth measurements. The studies show that the proposed techniques are valid and effective, as the estimation schemes adopted either fulfill the requirements imposed or improve the overall estimation performance. Highlighted results presented in this dissertation include a scale factor calibration accuracy of 286 ppm for a MEMS gyroscope (Chapter 3), an improvement of 15.1% of angular displacement estimation accuracy by adopting a threshold sensing technique for a scanning micro-mirror (Chapter 4), and a phase shift prediction error of 0.39 degree for a electrostatic micro-scanner using shared electrodes for actuation and sensing (Chapter 5).