GA-based stable intelligent control for a class of underactuated mechanical systems

Issue title: Proceedings of the International Conference on Mechatronics and Information Technology (ICMIT 2005, ICMIT 2007)

Article type: Research Article

Authors: Liu, Diantong^{a; *} | Guo, Weiping^a

Affiliations: [a] Institute of Computer Science and Technology, Yantai University, Shandong Province, China

Correspondence: [*] Corresponding author. E-mail: [email protected]

Abstract: The nonlinear dynamic model of a class of underactuated mechanical systems (UMSs) in a potential field is built with the Lagrangian method and some system properties are analyzed such as the system passivity. A genetic algorithm (GA) based stable intelligent control algorithm is proposed for the class of UMSs combining fuzzy control and sliding mode control. The real-valued GA is used to adjust the parameters of stable intelligent controller to improve the system performance. The underactuated double-pendulum-type overhead crane (DPTOC) is used to validate the proposed control algorithm. The system is proved to be stabilized asymptotically to its equilibrium with sliding mode theory. Simulation results illustrate the system's complexity and the validity of proposed control algorithm under different conditions.

Keywords: Underactuated mechanical systems, stable intelligent control, real-valued genetic algorithm, overhead crane

DOI: 10.3233/JAE-2011-1343

Journal: International Journal of Applied Electromagnetics and Mechanics, vol. 36, no. 1-2, pp. 49-61, 2011