Race car trajectory determination by multi-source acoustic emission analysis

Abstract

This talk presents a signal processing methodology to extract racecar trajectory and engine speed information from a multiple microphone array placed at track side. Phase differences, as well as Doppler shifts of the engine acoustic emissions from an array of microphones, provide multiple sources from which a robust determination of the vehicle trajectory can be extracted. Furthermore, time-frequency analysis of the Doppler-corrected signals provide a rich source of information about the instantaneous engine speed. The combination of both methods provides valuable information about the vehicle and engine characteristics. The analysis method used in this work is applied to acoustic emission data recorded by a track-side array of four microphones. The data analyzed include synthetic validation and calibration data, as well as actual race data. Special attention was devoted to optimizing the microphone array configuration for maximum resolution given an array size. By carefully exploiting the inherent redundancies in the data, a robust and computationally efficient analysis method was successfully developed. The results demonstrate that valuable vehicle and engine parameters can be extracted from such an approach.