Design of Real-Time Acquisition and Filtering for MEMS-based Accelerometer Data in Microcontroller

Abstract

Micro-electro-mechanical systems (MEMS) devices such as accelerometers, gyroscopes have now become indispensable sensors for automobile usages as well as for guidance subsystems such as inertial navigational units, etc. for aerial moving platforms. This paper presents real-time filtering of noise in the acquired acceleration data by the MEMS devices such as ADXL345. The MEMS accelerometer sensor is fitted on a motorized platform for acquiring noisy acceleration data in the x-y coordinates by running the platform on uneven surfaces. The novelty in our approach is that we have used the same microcontroller that is mounted on the motorized platform for acquiring the acceleration data as well as for filtering the noisy data in real-time. The finite impulse response (FIR) filter used for removal of noise is designed using Filter Design & Analysis (FDA) tool in MATLAB. Results of filtering noisy data in the microcontroller are matched with the results obtained by offline filtering done in MATLAB environment.