EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Reinforcement Learning‐Based Remote Human–Machine Shared Control Using Fuzzy Two‐Layer TD3

Gang Li, Xiangyu Guo, Shengnan Gao, Sihua Zhang, Zhe Dong, Zhonghua Pang

IET Control Theory & Applications · 2026-01 · 20 (1)

DOI: 10.1049/cth2.70196 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IET Control Theory & Applications, online erschienen am 05.11.2026. Verfasst von Gang Li und 5 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Key Technologies Research and Development Program, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Network delays pose a significant constraint on system performance in remote driving. Human–machine shared control offers a viable approach to mitigating its adverse effects by synergizing human driver experience with the inherent capabilities of vehicles in autonomous analysis and decision‐making. First, a remote driving human–machine shared control framework (RDHMSCF) comprising a network delay model, a driver model, a two‐degree‐of‐freedom vehicle dynamics model and a trajectory error model is established. Based on the twin delayed deep deterministic policy gradient (TD3), this paper proposes a fuzzy two‐layer TD3 (F‐TLTD3) algorithm for the dynamic allocation of human–machine shared control authority in RDHMSCF. To account for the impact of network delays and vehicle states, an error model‐based model predictive controller (MPC) and an F‐TLTD3‐based network delay‐aware human–machine shared control method are developed. Finally, performance metric functions are constructed to evaluate the control performance of the remote driving system. Based on the co‐simulation results from CarSim‐Simulink, the effectiveness of the F‐TLTD3 algorithm is verified.

Förderung

  • Key Technologies Research and Development Program
  • National Natural Science Foundation of China

Zitieren

APA

Li, G., Guo, X., Gao, S., Zhang, S., Dong, Z., & Pang, Z. (2026). Reinforcement Learning‐Based Remote Human–Machine Shared Control Using Fuzzy Two‐Layer TD3. IET Control Theory & Applications, 20(1). https://doi.org/10.1049/cth2.70196

Vancouver

Li G, Guo X, Gao S, Zhang S, Dong Z, Pang Z. Reinforcement Learning‐Based Remote Human–Machine Shared Control Using Fuzzy Two‐Layer TD3. IET Control Theory & Applications. 2026;20(1). doi:10.1049/cth2.70196

BibTeX
@article{Li2026,
  author = {Li, Gang and Guo, Xiangyu and Gao, Shengnan and Zhang, Sihua and Dong, Zhe and Pang, Zhonghua},
  title = {Reinforcement Learning‐Based Remote Human–Machine Shared Control Using Fuzzy Two‐Layer TD3},
  journal = {IET Control Theory & Applications},
  year = {2026},
  volume = {20},
  number = {1},
  doi = {10.1049/cth2.70196}
}
Kostenlos starten →