ANALYZING VERTICAL VIBRATIONS OF AUTOMOBILE WHEEL HUB TO MONITOR TYRE PRESSURE USING STATISTICAL FEATURES AND SUPPORT VECTOR MACHINE ALGORITHM

Abstract

One of the main safety measures used in automobiles are Tyre pressure monitoring systems (TPMS). These are intelligent devices fabricate to supervise the tyre pressure in automobile. The current technology use barometric sensors or vehicle speed sensors to measure the pressure directly. They mainly depend on batteries and different types of remote sensors which would increase the installation cost and complication maintenance. This paper suggests a novel technique adopting machine learning and fault diagnosis to supervise the vehicle tyre pressure indirectly. Vertical vibrations from a wheel hub are acquired using a three axis mems accelerometer sensor. After feature extraction and feature selection the selected features are classified using support vector machine algorithm. A good classification accuracy of 90% was gained.