Threshold value-based detection of relevant force inputs onto vehicle skin panels with piezoelectric signals

This article describes and compares autonomous threshold value-based processes for detecting force inputs onto vehicle skin panels. For this purpose, 13 piezoelectric foil sensors made from polyvinylidene fluoride are applied to the inside of the outer skin panels as sensors. Applied forces give rise to mechanical vibrations in the material, which, in the form of expansions in the sensor, result in a proportional output signal. On the basis of these signals, two static and two adaptive threshold value processes are presented and evaluated for differentiating between relevant events (including scratches, parking bumps) and irrelevant events (including wind, rain). At the same time, the central issue of optimum configuration of the threshold value is investigated, and solutions proposed.

► We show a suitable sensor and corresponding signal processing for detecting force impacts onto vehicle skin panels.
► We present different algorithms for detecting relevant force impacts.
► Based on simulation we optimise the algorithms to have a low false detection and rejection rate.
► A comparison of the algorithms shows that an adaptive bipolar detection mechanism is the best.
► The results and algorithms itself can be used in other signal detection applications.