Active Reduction of Transient Driveline Oscillations with Fuzzy Update Timings of Control Input

In active vibration suppression for automotive drivelines, this research presents an effective fuzzy inference compensation for the time-varying long control cycle limitation. First, a simplified driveline model with the control cycle limitation is provided for investigating active damping strategy. As a baseline damping strategy, a sampled-data H
∞
controller is used to mitigate effects of discretization errors due to the long control cycles. In addition, the control cycles force the control input updating timing of the actuator to be perturbed from that of the fixed periodical controller. The core idea of the compensation is to regard these uncertain timings as fuzzy sets such as "Approximately past timing" and "Approximately future timing". The proposed fuzzy logic simply computes feasible damping commands at various timings based on Mamdani-type reasoning with a few rules. Through simulation tests with various situations, the superior robustness of the proposed method is confirmed.