Neural network adaptive sliding mode control for omnidirectional vehicle with uncertainties

Xingyang Lu; Xiangyin Zhang; Guoliang Zhang; Jinhui Fan; Songmin Jia

doi:10.1016/j.isatra.2018.10.043

Neural network adaptive sliding mode control for omnidirectional vehicle with uncertainties

ISA Trans. 2019 Mar:86:201-214. doi: 10.1016/j.isatra.2018.10.043. Epub 2018 Nov 3.

Authors

Xingyang Lu¹, Xiangyin Zhang², Guoliang Zhang³, Jinhui Fan⁴, Songmin Jia¹

Affiliations

¹ Faulty of Information Technology, Beijing University of Technology, Beijing 100124, China; Beijing Key Laboratory of Computational Intelligence and Intelligent System, Beijing 100124, China; Engineering Research Center of Digital Community, Ministry of Education, Beijing 100124, China.
² Faulty of Information Technology, Beijing University of Technology, Beijing 100124, China; Beijing Key Laboratory of Computational Intelligence and Intelligent System, Beijing 100124, China. Electronic address: zhangxgyn@foxmail.com.
³ Faulty of Information Technology, Beijing University of Technology, Beijing 100124, China; Beijing Key Laboratory of Computational Intelligence and Intelligent System, Beijing 100124, China.
⁴ Faulty of Information Technology, Beijing University of Technology, Beijing 100124, China; Engineering Research Center of Digital Community, Ministry of Education, Beijing 100124, China.

PMID: 30420141
DOI: 10.1016/j.isatra.2018.10.043

Abstract

This paper presents a novel neural network adaptive sliding mode control (NNASMC) method to design the dynamic control system for an omnidirectional vehicle. The omnidirectional vehicle is equipped with four Mecanum wheels that are actuated by separate motors, and thus has the omnidirectional mobility and excellent athletic ability in a narrow space. Considering various uncertainties and unknown external disturbances, kinematic and dynamic models of the omnidirectional vehicle are established. The inner-loop controller is designed based the sliding mode control (SMC) method, while the out-loop controller uses the proportion integral derivative (PID) method. In order to achieve the stable and robust performance, the artificial neural network (ANN) based adaptive law is introduced to model and estimated the various uncertainties disturbances. Stability and robustness of the proposed control method are analyzed using the Lyapunov theory. The performance of the proposed NNASMC method is verified and compared with the classical PID controller and SMC controller through both the computer simulation and the platform experiment. Results validate the effectiveness and robustness of the NNASMC method in presence of uncertainties and unknown external disturbances.

Keywords: Artificial neural network; Mecanum wheel; Omnidirectional vehicle; Sliding mode control.