A High-Order Tracking Differentiator for Improved Filtering and Derivative Estimation

Shanhai Jin,Xinyue Tian,Dejin Zhao,Xiaodan Wang,Changfu Jin
DOI: https://doi.org/10.1109/ICAR58858.2023.10406727
2023-01-01
Abstract:The task of filtering and estimating derivatives from noisy signals is crucial in feedback control systems. While linear filtering methods are widely used due to their simplicity, they can introduce significant phase lag and amplify noise, negatively impacting system performance. To address these limitations, researchers have investigated tracking differentiators (TD) based on sliding mode techniques. However, existing TD methods are restricted to estimating first-order or arbitrary order derivatives with significant overshoot during the process. To overcome these challenges, this paper proposes a novel highorder sliding mode tracking differentiator (TD-N) with gains adjusted using a nested signum function to provide reliable high-order derivative estimation without excessive overshoot. The simulation results confirm the effectiveness and advantages of the proposed TD-N.
What problem does this paper attempt to address?