Event-Based Sampling for Handwheel Vibration Analysis

Over the past 20 years automobile quality, and consumer perceptions thereof, have been an increasing part of the vehicle engineering process. In automotive systems, torque-induced handwheel vibration is a significant source of consumer annoyance.

When dealing with torque recordings containing a single or very small number of sinusoids, a time-referenced signal can easily be manipulated to yield instantaneous torque estimates. As the number of sinusoids (orders) grows, however, this becomes more difficult.

In this paper, the authors demonstrate a simple technique to effectively estimate instantaneous torque level of high-order torque components when there are many higher-order sinusoids. The point is made that basic signal processing techniques applicable to time-referenced signals are also applicable to those referenced to angle or other variables.