文献类型：期刊文献

英文题名：Adaptive self-structuring neural network tracking control for underactuated USVs with actuator faults and input saturation

作者：Liu, Haitao[1,2,3];Zhou, Xuecheng[1,2];Tian, Xuehong[1,2,3];Mai, Qingqun[1,2,3];Li, Ronghui[3]

机构：[1]Guangdong Ocean Univ, Sch Mech Engn, Zhanjiang 524088, Peoples R China;[2]Guangdong Ocean Univ, Shenzhen Inst, Shenzhen 518120, Peoples R China;[3]Guangdong Engn Technol Res Ctr Ocean Equipment & M, Guangdong Prov Key Lab Intelligent Equipment South, Zhanjiang 524088, Peoples R China

年份：2024

卷号：309

外文期刊名：OCEAN ENGINEERING

收录：SCI-EXPANDED(收录号：WOS:001258749800001)、、EI(收录号：20242516295534)、Scopus(收录号：2-s2.0-85196374869)、WOS

基金：This work was supported by the Key Project of the Department of Education of Guangdong Province [grant numbers 2023ZDZX1005, 2021ZDZX1041], Shenzhen Science and Technology Program [grant number JCYJ20220530162014033], Guangdong Basic and Applied Basic Research Foundation [grant number 2024A1515011345], National Natural Science Foundation of China [grant numbers 52171346, 62171143], and the Science and Technology Planning Project of Zhanjiang City [grant numbers 2021A05023 and 2021E05012].

语种：英文

外文关键词：Underactuated unmanned surface vessels; Predefined-time stable; Error constraints; Self-structuring NNs; Input saturation; Fault-tolerant control

外文摘要：In this paper, a predefined-time trajectory tracking fault-tolerant controller based on backstepping is proposed. To solve the problem of large control inputs of underactuated unmanned surface vehicles (USVs), a predefinedtime auxiliary dynamics system that combines a new saturation function is constructed to compensate for the effects of input saturation and smooth the control inputs. To make the error-constrained USVs unconstrained by the initial value of the performance function, the error conversion function is combined with the barrier Lyapunov function (BLF) so that the movement of the USV is no longer affected by the initial conditions. To address the effects of model uncertainty, external environmental perturbations, and actuator faults on the tracking performance, self-structuring neural networks (SSNNs) are used to approximate complex unknown nonlinear functions, and the number of neurons can be adjusted. Finally, a stability analysis and numerical simulations show that the closed-loop system is predefined-time stable and the control inputs are limited to the set range, and the SSNNs can approximate uncertain terms well, which improves the system robustness and reduces the amount of computation. The initial position conditions do not limit the operation of the USVs, and all errors within the closed-loop system converge in a predefined time.