文献类型：期刊文献

英文题名：Augmented model-based dynamic positioning predictive control for underactuated unmanned surface vessels with dual-propellers

作者：Gao, Shuang[1];Liu, Chunping[1];Tuo, Yulong[2];Chen, Kun[1];Zhang, Tong[1]

机构：[1]Shaoxing Univ, Sch Mech & Elect Engn, Shaoxing 312000, Peoples R China;[2]Dalian Maritime Univ, Coll Marine Elect Engn, Dalian 116000, Peoples R China

年份：2022

卷号：266

外文期刊名：OCEAN ENGINEERING

收录：SCI-EXPANDED(收录号：WOS:000882777500002)、、EI(收录号：20224212974477)、Scopus(收录号：2-s2.0-85139820263)、WOS

基金：The supports of the National Natural Science Foundation of China (Grant No. 52101298) and the Natural Science Foundation of Zhejiang Province, China (Grant No. LQ22E090007) are gratefully acknowledged.

语种：英文

外文关键词：Dynamic positioning; Underactuated unmanned surface vessel; Model predictive control; Model augmentation; Disturbance observer

外文摘要：The dynamic positioning control for underactuated unmanned surface vessel (u-USV) is difficult but urgently needed. This paper aims to propose an effective dynamic positioning controller for u-USV. The accuracy of the internal dynamics of u-USV is improved by designing an augmented model (AM) based on a residual neural network. After that, an AM-based finite-time observer (AM-FTO) is designed. The disturbance observed by the AM-FTO is mainly the environmental disturbance, not the original composite disturbance anymore. Further, the direction of the environmental disturbance can be estimated. Taking the direction as the desired heading, a dynamic positioning controller is designed for u-USV based on the nonlinear model predictive control (NMPC) method. Simulations and experiments show that the proposed method can track the internal dynamics of u-USV with high accuracy. Additionally, the designed dynamic positioning controller has better performance than the existing popular NMPC method.