Del Rio Asier, Barambones Oscar, Artetxe Eneko, Uralde Jokin, Calvo Isidro
Department of Systems Engineering and Automatic Control, Faculty of Engineering of Vitoria-Gasteiz, University of the Basque Country (UPV/EHU), 01006 Vitoria-Gasteiz, Spain.
Micromachines (Basel). 2025 Apr 15;16(4):469. doi: 10.3390/mi16040469.
Piezoelectric actuators (PEAs) are widely used in high-precision applications but suffer from nonlinear hysteresis effects that degrade positioning accuracy. To address this challenge, this study presents a backstepping controller with an Artificial Neural Network (ANN)-based feedforward compensation scheme to enhance trajectory tracking performance. The ANN compensates for the hysteresis effects, while the backstepping strategy ensures robust reference tracking. The proposed controller is validated through real-time experiments using a piezoelectric actuator system. Comparative analysis with a conventional PID controller demonstrates the superiority of the backstepping approach, achieving significantly lower tracking errors across different reference signals and frequencies. Error metrics have been employed to confirm the improved accuracy and robustness of the proposed method. These findings highlight the effectiveness of the proposed ANN-enhanced backstepping control in overcoming hysteresis-related challenges in precision positioning applications.
压电致动器（PEA）广泛应用于高精度应用中，但存在非线性滞后效应，会降低定位精度。为应对这一挑战，本研究提出了一种基于人工神经网络（ANN）前馈补偿方案的反步控制器，以提高轨迹跟踪性能。人工神经网络补偿滞后效应，而反步策略确保稳健的参考跟踪。通过使用压电致动器系统的实时实验对所提出的控制器进行了验证。与传统PID控制器的对比分析表明了反步方法的优越性，在不同参考信号和频率下实现了显著更低的跟踪误差。已采用误差指标来确认所提方法提高的精度和鲁棒性。这些发现突出了所提人工神经网络增强反步控制在克服精密定位应用中与滞后相关挑战方面的有效性。
