EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Robust Bipedal Locomotion via Stein Variational Gradient Descent Solution Framework for Model Predictive Control

Qing Yang, Xian-Qian Hong, Xu Wang, Xin-Long Yu, Bin Lan

Electronics · 05.03.2026 · 15 (5) · 1083

DOI: 10.3390/electronics15051083 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 05.03.2026. Verfasst von Qing Yang und 4 weiteren Autor:innen. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Anhui Provincial Natural Science Foundation, Anhui Provincial Key Laboratory of Humanoid Robots, Anhui Provincial Industry Innovation Center of Humanoid Robots.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper proposes a Stein Variational Gradient Descent (SVGD) solution framework for Model Predictive Control (MPC) to enhance robustness in bipedal locomotion. While conventional simplified rigid-body dynamics (SRBD) based MPC with deterministic solvers such as the active-set method performs reliably during straight-line walking, its performance degrades significantly in agile maneuvers, where modeling simplifications become more pronounced. Simulations on the 43-DoF QingLoong humanoid demonstrate that SVGD-MPC exhibits improved robustness in dynamic locomotion tasks, including high-angular-velocity turning and circular walking. However, conventional MPC fails to maintain stability. Rather than relying on a single nominal solution, SVGD-MPC maintains an ensemble of candidate control trajectories that evolve jointly within the MPC optimization process. This ensemble-based formulation reduces sensitivity to modeling inaccuracies and allows the controller to retain feasible control candidates when the assumptions underlying the SRBD model are violated. The method sacrifices marginal tracking precision to maintain viability under model inadequacy. Moreover, our framework establishes a paradigm for robust locomotion control where the traditional optimization method fails to work.

Förderung

  • Anhui Provincial Natural Science Foundation
  • Anhui Provincial Key Laboratory of Humanoid Robots
  • Anhui Provincial Industry Innovation Center of Humanoid Robots

Zitieren

APA

Yang, Q., Hong, X., Wang, X., Yu, X., & Lan, B. (2026). Robust Bipedal Locomotion via Stein Variational Gradient Descent Solution Framework for Model Predictive Control. Electronics, 15(5), 1083. https://doi.org/10.3390/electronics15051083

Vancouver

Yang Q, Hong X, Wang X, Yu X, Lan B. Robust Bipedal Locomotion via Stein Variational Gradient Descent Solution Framework for Model Predictive Control. Electronics. 2026;15(5):1083. doi:10.3390/electronics15051083

BibTeX
@article{Yang2026,
  author = {Yang, Qing and Hong, Xian-Qian and Wang, Xu and Yu, Xin-Long and Lan, Bin},
  title = {Robust Bipedal Locomotion via Stein Variational Gradient Descent Solution Framework for Model Predictive Control},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {5},
  pages = {1083},
  doi = {10.3390/electronics15051083}
}
Kostenlos starten →