Adaptive neural network control of uncertain MIMO nonlinear systems with input saturation. Issue 5 (July 2016)

• Record Type: Journal Article
• Authors: Zhou, Shengfeng; Chen, Mou; Ong, Chong-Jin; Chen, Peter
• Is Part Of: Neural computing & applications. Volume 27:Issue 5(2016)
• Page Start: 1317
• Page End: 1325
• Publication Date: 2016-07
• Subjects: Neural networks (NNs) -- Input saturation -- MIMO systems -- Adaptive tracking control; Neural networks (Computer science) -- Periodicals; Neural circuitry -- Periodicals; Artificial intelligence -- Periodicals; Neural Networks (Computer) -- Periodicals; Réseaux neuronaux (Informatique) -- Périodiques; Réseaux nerveux -- Périodiques; Intelligence artificielle -- Périodiques; 006.32
• DOI: 10.1007/s00521-015-1935-7
• Languages: English
• ISSNs: 0941-0643

Abstract

Abstract In this paper, an adaptive neural network (NN) tracking controller is developed for a class of uncertain multi-input multi-output (MIMO) nonlinear systems with input saturation. Radial basis function neural networks are utilized to approximate the unknown nonlinear functions in the MIMO system. A novel auxiliary system is developed to compensate the effects induced by input saturation (in both magnitude and rate) during tracking control. Endowed with a switching structure that integrates two existing representative auxiliary system designs, this novel auxiliary system improves control performance by preserving their advantages. It provides a comprehensive design structure in which parameters can be adjusted to meet the required control performance. The auxiliary system signal is utilized in both the control law and the neural network weight-update laws. The performance of the resultant closed-loop system is analyzed, and the bound of the transient error is established. Numerical simulations are presented to demonstrate the effectiveness of the proposed adaptive neural network control.