EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Genetic Algorithm Optimized Sliding Mode Control for 6-DOF Commercial Vehicle Piezoelectric Active Suspension with RBF Neural Network Compensation

Junbiao Xie, Yuying Jiang, Chen Wang, Jingcheng Dai, Yiming Yu, Chenglong Pan

Vibration · 26.05.2026 · 9 (2) · 38

DOI: 10.3390/vibration9020038 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Vibration, online erschienen am 26.05.2026. Verfasst von Junbiao Xie und 5 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shandong Provincial Natural Science Foundation, Shandong Provincial College Students’ Innovation and Entrepreneurship Training Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To address the vibration reduction problem of the six-degrees of freedom(6-DOF) half-vehicle model and to improve ride comfort and handling stability, a piezoelectric stack actuator based on the inverse piezoelectric effect was introduced. A 6-DOF half-vehicle dynamic model coupling the cab, body, and wheels was established based on the Lagrange equation. Based on this model, a vertical-pitch dual sliding surface RBF neural network sliding mode control strategy was proposed, with two independent RBF neural networks designed to separately approximate, online, the comprehensive uncertainties in the vertical and pitch channels associated with unmodeled dynamics, external disturbances, and modeling simplifications. The variable-speed reaching law (dsat) function was used to design the sliding mode reaching law, balancing sliding surface convergence speed and vibration suppression. Six indicators, including vertical acceleration of the cab and vertical acceleration of the vehicle body, were selected as performance evaluation metrics to establish the fitness function. Combined with a genetic algorithm, the dual sliding surface coefficients, RBF network parameters, adaptive update rates, and variable-speed reaching law parameters were globally optimized. The vibration reduction effects of four schemes—passive control, traditional sliding mode control, RBF sliding mode control, and genetic algorithm optimized RBF dual-sliding-mode control—were compared and analyzed. Simulation results show that the genetic algorithm optimized RBF dual-sliding-mode control achieves improved vibration suppression in several key ride-comfort-related indices and provides better overall coordination among ride comfort, suspension working space, and tire dynamic deflection. The research results validate the effectiveness of this method and provide a new solution for addressing vehicle vibration reduction problems.

Förderung

  • Shandong Provincial Natural Science Foundation
  • Shandong Provincial College Students’ Innovation and Entrepreneurship Training Program

Zitieren

APA

Xie, J., Jiang, Y., Wang, C., Dai, J., Yu, Y., & Pan, C. (2026). Genetic Algorithm Optimized Sliding Mode Control for 6-DOF Commercial Vehicle Piezoelectric Active Suspension with RBF Neural Network Compensation. Vibration, 9(2), 38. https://doi.org/10.3390/vibration9020038

Vancouver

Xie J, Jiang Y, Wang C, Dai J, Yu Y, Pan C. Genetic Algorithm Optimized Sliding Mode Control for 6-DOF Commercial Vehicle Piezoelectric Active Suspension with RBF Neural Network Compensation. Vibration. 2026;9(2):38. doi:10.3390/vibration9020038

BibTeX
@article{Xie2026,
  author = {Xie, Junbiao and Jiang, Yuying and Wang, Chen and Dai, Jingcheng and Yu, Yiming and Pan, Chenglong},
  title = {Genetic Algorithm Optimized Sliding Mode Control for 6-DOF Commercial Vehicle Piezoelectric Active Suspension with RBF Neural Network Compensation},
  journal = {Vibration},
  year = {2026},
  volume = {9},
  number = {2},
  pages = {38},
  doi = {10.3390/vibration9020038}
}
Kostenlos starten →