Title: Adaptive dynamic fuzzy neural network-based decoupled sliding-mode controller with hybrid sliding surfaces

Authors: Guoliang Zhao; Hongxing Li; Zhankui Song

Addresses: Faculty of Electronic Information and Electronic Engineering, Dalian University of Technology, Dalian 116024, China; Modern Manufacture Engineering Centre, Heilongjiang University of Science and Technology, Harbin, 150022, China ' Faculty of Electronic Information and Electronic Engineering, Dalian University of Technology, Dalian 116024, China ' Faculty of Electronic Information and Electronic Engineering, Dalian University of Technology, Dalian 116024, China

Abstract: The main motivation for this study is to remove the constraint of known non-linear dynamics in the control law, and eliminate the chattering in the switching. Thus, the dynamic fuzzy neural network (DFNN) is applied to better approximate the unknown non-linear dynamics in a decoupled sliding mode control (DSMC) problem for a class of fourth-order non-linear systems. To avoid the drawback of control chattering occurring in the sliding mode, a fractional order PI compensator is used to compensate for disturbance forces while suppressing chattering effect. Using this approach, the response of the system will converge faster than that of previous reports. Finally, simulations adopting this framework are presented for cart-pendulum system and TORA system. Simulation results show the effectiveness of the newly proposed adaptive DFNN-based decoupled sliding mode control method.

Keywords: decoupled SMC; sliding mode control; DSMC; fractional order PI controllers; adaptive control; dynamic fuzzy neural networks; DFNN; hybrid sliding surfaces; nonlinear dynamics; switching; control chattering; cart-pendulum system; TORA system; simulation.

DOI: 10.1504/IJAAC.2013.057058

International Journal of Automation and Control, 2013 Vol.7 No.3, pp.183 - 201

Received: 24 Jul 2013
Accepted: 01 Aug 2013
Published online: 12 Jul 2014 *

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