Data-driven H∞ vibration control design and verification for an active suspension system with unknown pseudo-drift dynamics.

This paper presents a data-based vibration control algorithm that can realize the vibration suppression of a laboratory suspension facility without knowing its drift dynamics. The vibration controller consists of two sub-controllers, a prestable controller that assists in generating higher quality operational data and a data-driven H∞ vibration controller that realizes vibration suppression. The proposed control algorithm only needs to know partial knowledge of the system dynamics without any information about the so-called pseudo-drift dynamics. We analyse and evaluate the control approach through simulation and experimentation based on a laboratory suspension platform. The comparison results shows that the data-based H∞ vibration control is quite promising. In addition, the influences of two critical parameters on the data-based controller are quantitatively studied, which can be used to optimize the control structure better and improve control performance.