Efficient input shaping control for flexible structures with unscented Kalman filter

With the advancement of the industry, increasing attention has been directed toward residual vibration issues in flexible structures and underactuated systems. Input shaping has been demonstrated as an effective vibration suppression technique. However, conventional input shapers often face challenges associated with parameter inaccuracies, resulting in suboptimal control performance. To address these critical limitations, this paper proposes an Unscented Kalman filter-based Zero vibration derivative Input shaping (UZI) method with twofold ideas: a) employing a data-driven Unscented Kalman Filter for system parameter identification to enhance control efficiency, and b) suppressing residual vibrations through a zero vibration derivative input shaper. For performance validation, an authentic dataset of vertical flexible beams is established and publicly released. Empirical studies demonstrate that the proposed UZI achieves remarkable vibration suppression effects compared with several state-of-the-art approaches.