Title: A novel robust online sustainable adaptation dynamics control method for robot movement by wheel type

Authors: Nguyen Minh Quang; Le Thi Phuong Thanh; Nguyen Tien Tung

Addresses: School of Mechanical and Automotive Engineering, Hanoi University of Industry, 298 Cau Dien, Bac Tu Liem District, Ha Noi, 084-100000, Vietnam ' School of Mechanical and Automotive Engineering, Hanoi University of Industry, 298 Cau Dien, Bac Tu Liem District, Ha Noi, 084-100000, Vietnam ' School of Mechanical and Automotive Engineering, Hanoi University of Industry, 298 Cau Dien, Bac Tu Liem District, Ha Noi, 084-100000, Vietnam

Abstract: A novel robust control algorithm called reinforcement learning online sustainable adaptive dynamic control (RLOSADC) was developed to solve the problem of approximation for nonlinear systems with absolutely no information about internal dynamics. The proposed control model was built based on a new algorithm called optimise cooperation many nonlinear systems (OCMNO) with powerful features and convergence capabilities. New and unique features of the proposed model are shown through a highly flexible design and control procedure. The traditional robot dynamic model is transformed into a tight feedback nonlinear system model for designing 'integrated' kinetic and dynamic control laws to overcome the disadvantages of the previous method. The RLOSADC model has been applied to cling control robust, sustainable adaptation for optimising the kinematic and dynamic clinging quality indicators for robot movement by wheel type (RMWT). Numerical and experimental simulation results on RMWT show the effectiveness of the proposed RLOSADC control model.

Keywords: OCMNO; RLOSADC; reinforcement learning online sustainable adaptive dynamic control; RMWT; robot movement by wheel type; adaptive control; nonlinear system.

DOI: 10.1504/IJMMS.2023.137367

International Journal of Mechatronics and Manufacturing Systems, 2023 Vol.16 No.4, pp.339 - 363

Received: 03 Dec 2022
Accepted: 13 Jun 2023
Published online: 14 Mar 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article