EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Trajectory Tracking Control for Robotic Manipulator Based on Soft Actor–Critic and Generative Adversarial Imitation Learning

Jintao Hu, Fujie Wang, Xing Li, Yi Qin iD, Fang Guo, Ming Jiang

Biomimetics · 21.12.2024 · 9 (12) · 779

DOI: 10.3390/biomimetics9120779 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biomimetics, online erschienen am 21.12.2024. Verfasst von Jintao Hu und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science and Technology, Dongguan University of Technology, Dongguan 523808, China. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, GuangDong Basic and Applied Basic Research Foundation, Characteristic Innovation Foundation of Guangdong Education Department.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In this paper, a deep reinforcement learning (DRL) approach based on generative adversarial imitation learning (GAIL) and long short-term memory (LSTM) is proposed to resolve tracking control problems for robotic manipulators with saturation constraints and random disturbances, without learning the dynamic and kinematic model of the manipulator. Specifically, it limits the torque and joint angle to a certain range. Firstly, in order to cope with the instability problem during training and obtain a stability policy, soft actor–critic (SAC) and LSTM are combined. The changing trends of joint position over time are more comprehensively captured and understood by employing an LSTM architecture designed for robotic manipulator systems, thereby reducing instability during the training of robotic manipulators for tracking control tasks. Secondly, the obtained policy by SAC-LSTM is used as expert data for GAIL to learn a better control policy. This SAC-LSTM-GAIL (SL-GAIL) algorithm does not need to spend time exploring unknown environments and directly learns the control strategy from stable expert data. Finally, it is demonstrated by the simulation results that the end effector of the robot tracking task is effectively accomplished by the proposed SL-GAIL algorithm, and more superior stability is exhibited in a test environment with interference compared with other algorithms.

Schlagwörter

Stichwörter der Autor:innen

deep reinforcement learningend effectorgenerative adversarial imitation learninglong short-term memoryrobotics tracking control

Autor:innen und Institutionen

  • School of Computer Science and Technology, Dongguan University of Technology, Dongguan 523808, China.

ORCID-Profile: Yi Qin

Förderung

  • National Natural Science Foundation of China
  • GuangDong Basic and Applied Basic Research Foundation
  • Characteristic Innovation Foundation of Guangdong Education Department
  • Liaoning Natural Science Foundation
  • Dongguan Science and Technology of Social Development Program
  • SSL Sci-tech Commissioner Program

Zitieren

APA

Hu, J., Wang, F., Li, X., Qin, Y., Guo, F., & Jiang, M. (2024). Trajectory Tracking Control for Robotic Manipulator Based on Soft Actor–Critic and Generative Adversarial Imitation Learning. Biomimetics, 9(12), 779. https://doi.org/10.3390/biomimetics9120779

Vancouver

Hu J, Wang F, Li X, Qin Y, Guo F, Jiang M. Trajectory Tracking Control for Robotic Manipulator Based on Soft Actor–Critic and Generative Adversarial Imitation Learning. Biomimetics. 2024;9(12):779. doi:10.3390/biomimetics9120779

BibTeX
@article{Hu2024,
  author = {Hu, Jintao and Wang, Fujie and Li, Xing and Qin, Yi and Guo, Fang and Jiang, Ming},
  title = {Trajectory Tracking Control for Robotic Manipulator Based on Soft Actor–Critic and Generative Adversarial Imitation Learning},
  journal = {Biomimetics},
  year = {2024},
  volume = {9},
  number = {12},
  pages = {779},
  doi = {10.3390/biomimetics9120779}
}
Kostenlos starten →