loading page

Emergency supplies transportation robot trajectory tracking control based on Koopman and improved event-triggered model predictive control
  • +4
  • Yaqi Zhang,
  • Minan Tang,
  • Haiyan Zhang,
  • Bo An,
  • Yaguang Yan,
  • Wenjuan Wang,
  • Kunxi Tang
Yaqi Zhang
Lanzhou Jiaotong University
Author Profile
Minan Tang
Lanzhou Jiaotong University

Corresponding Author:tangminan@mail.lzjtu.cn

Author Profile
Haiyan Zhang
Jinan New and Old Kinetic Energy Conversion Starting Area Management Committee
Author Profile
Bo An
Lanzhou Jiaotong University
Author Profile
Yaguang Yan
Lanzhou Jiaotong University
Author Profile
Wenjuan Wang
Lanzhou Jiaotong University
Author Profile
Kunxi Tang
Lanzhou Jiaotong University
Author Profile

Abstract

In the emergency rescue and disposal of social public emergencies, supply transportation effectively provides a strong supply foundation and realistic conditions. The trajectory tracking control of emergency supplies transportation robot is the key technology to ensure the timeliness of transportation. In this paper, the emergency supplies transportation robot is taken as the research object, based on Koopman operator theory, combined with radial basis function (RBF) neural network disturbance observer and adaptive prediction horizon event-triggered model predictive control (APET-MPC) algorithm to investigate the purely data-driven trajectory tracking control problem of emergency supplies transportation robot when the model parameters and models are unknown. Firstly, the Koopman operator is used to establish a high-dimensional linear model of the robot. Secondly, the RBF neural network disturbance observer is designed to estimate the disturbance during the robot operation and compensate it to the controller. Thirdly, APET-MPC is used to optimize the trajectory tracking control of the emergency supplies transportation robot to reduce computational complexity. Finally, the performance of the proposed trajectory tracking controller is verified by Carsim/ Simulink joint simulation. The simulation results show that the model established by Koopman operator theory can achieve the high accuracy approximation of the robot. Compared with the MPC trajectory tracking controller, the APET-MPC trajectory tracking controller based on RBF neural network disturbance observer (RBF-APET-MPC) improves the tracking accuracy of the robot and reduces the total triggering times of the system by more than 50%.
29 Apr 2023Submitted to International Journal of Robust and Nonlinear Control
29 Apr 2023Submission Checks Completed
29 Apr 2023Assigned to Editor
29 Apr 2023Review(s) Completed, Editorial Evaluation Pending
12 May 2023Reviewer(s) Assigned
26 Oct 2023Editorial Decision: Revise Minor
21 Nov 20231st Revision Received
30 Jan 2024Review(s) Completed, Editorial Evaluation Pending
01 Feb 2024Editorial Decision: Revise Minor
21 Feb 20243rd Revision Received
21 Feb 2024Review(s) Completed, Editorial Evaluation Pending
22 Feb 2024Reviewer(s) Assigned
11 May 2024Editorial Decision: Accept