未知动态海况下无人水面艇的鲁棒协同路径跟踪研究

项目名称： 未知动态海况下无人水面艇的鲁棒协同路径跟踪研究

项目编号： No.51209026

项目类型： 青年科学基金项目

立项/批准年度： 2013

项目学科： 水利科学与海洋工程学科

项目作者： 彭周华

作者单位： 大连海事大学

项目金额： 25万元

中文摘要： 无人水面艇由于小型轻量、操纵灵活、自主性强和不必冒人员生命危险等优点在军事和民用领域具有广泛的应用前景。无人水面艇合作控制能够满足多无人水面艇协同执行任务的需求。国内外对无人水面艇合作控制理论研究十分有限，缺乏系统化的设计方法。本课题将探索无人水面艇合作控制中尚未解决的关键理论和控制难题，重点研究未知动态海洋环境下欠驱动无人水面艇的鲁棒协同路径跟踪问题。以新型误差反馈滤波动态面为主要设计手段，综合应用路径跟踪原理、迭代学习原理与时变参数神经网络逼近原理进行协同控制器设计，结合李亚普诺夫理论、输入状态稳定、Lyapunov-Krasovskii函数以及图论进行闭环系统稳定性分析。目的在于通过深入的研究，建立系统而完善的无人水面艇合作控制理论，为无人水面艇合作控制的工程应用打下坚实的理论基础，使我国在无人水面艇合作控制方面的研究领先或同步于世界发展水平，同时培养船舶与海洋工程领域专业技术人才。

中文关键词： 无人水面艇；协同路径跟踪；神经网络；分散式控制；动态面

英文摘要： Unmanned surface vehicles (USVs) have broad applications both in military and civilian fields due to its small size, flexible manipulation, greater autonomy and capabilities without risking human life, etc. Cooperative control can meet the requirements of multiple USVs to perform collaborative tasks. However, the research on cooperative control theory of USVs is very limited both in China and overseas, and lacks a systematic design method. This project aims to investigate the key unsolved theoretical and technical problems on the design of cooperative control of USVs, and mainly focuses on the robust coordinated path following problem of underactuated USVs in the unknown dynamical marine environment. Based on the new developed error feedback filtering of dynamic surface control technique, the coordinated path following controllers are designed by incorporating the path following principles, iterative learning methods and approximation theory of time-varying neural networks. The stability analysis of the system is practiced with the aid of Lyapunov theory, input-to-state stability， Lyapunov-Krasovskii function, and graph theory. Through deeply research, this project will establish systematic and self-contained cooperative control theory of USVs and lay a solid theoretical foundation for its engineering applicatio

英文关键词： Unmanned Surface Vehicles；Coordinated Path Following；Neural Networks；Decentralized Control；Dynamic Surface Control