网络环境下船舶编队的分布式自适应控制

Based on the team's previous research on formation control of marine surface vehicles (MSVs) and the applicants' advanced research work in adaptive control for uncertain nonlinear systems, considering the relationship between the nonlinear dynamics of the MSV and topological characteristics of a network, this project aims to uncover the mechanisms of formation control of networked MSVs with uncertain dynamics and develop distributed coordinated control methods, in combination with individual control of single vehicle and distributed control of the group in a network environment. Drawing lessons from the distributed control methods of multi-agent systems, neural network based adaptive control and dynamic surface control techniques, this research will focus on the formation controller design methods and stability analysis with the aid of graph theory, matrix theory, singularity analysis and Lyapunov stability analysis. The objective of this research is to clarify the principles for formation control of networked MSVs based on multi-agent control techniques, establish a formation control framework for networked MSVs with uncertain dynamics, and solve the distributed adaptive formation control problem of networked MSVs. From a theoretical and practical standpoint, this research will enrich the control methods of networked nonlinear multi-agent systems, and provide the theoretical fundamentals for marine applications, such as automatic ocean resource exploration, environmental monitoring, underway replenishment, cooperative search and rescue, and so on. The simplicity, effectiveness, and robustness of the formation controllers are mainly targeted.

在课题组船舶编队控制研究的基础之上，充分发挥申请人在自适应控制解决非线性不确定系统控制问题方面所具有的技术优势，针对船舶个体非线性动力特性与网络拓扑特性之间的关系，将船舶运动的个体控制与网络环境下分布式群体控制相结合，重点研究含不确定动态的网络化船舶编队控制机理与分布式协调控制方法。以多智能体系统分布式控制、自适应神经网络控制和动态面控制为主要技术手段，综合应用图论、矩阵理论、奇异值分析、李亚普诺夫稳定性理论进行控制器设计与稳定性分析。目的在于探索基于多智能技术的网络化船舶编队控制原理，建立含不确定动态的网络化船舶编队控制框架，解决网络环境下船舶编队的分布式自适应控制问题。其理论意义和实际意义在于：理论上进一步丰富网络非线性多智能体系统的控制方法；应用上为利用多船舶编队进行海洋资源探测、环境监测、远洋补给、编队搜索救援提供控制理论基础。编队控制器的简捷、高效、鲁棒是课题研究的宗旨

多船舶编队控制在海洋资源探测、海洋环境监测、海洋结构物拖拽、海难搜索与营救等方面具有广阔的应用。项目针对船舶个体非线性动力特性与网络拓扑特性之间的关系，将船舶个体运动控制与网络环境下分布式群体控制相结合，提出了网络环境下多船舶编队的分布式控制理论框架；针对现有集中式编队控制方法的缺陷，提出了基于多智能体理论的分布式编队控制方法，设计了兼具鲁棒性和灵活性的无领航者和领航-跟随编队控制协议；将船舶动力学模型参数不确定、未建模动态以及海洋环境扰动融入到编队控制器设计，提出了基于神经网络的自适应动力学控制方法，提高了编队控制器的鲁棒性和适应性；针对传统编队控制器设计方法带来的计算复杂性问题，提出了动态面控制器设计方法进行多船舶编队控制器设计，使得编队控制算法简捷，并且有利于实际实现。.项目研究小组在IEEE Trans. on Neural Networks and Learning Systems, IEEE Trans. on Mechatronics, IEEE Trans. on Industrial Electronics, IEEE Trans. on Systems, Man, and Cybernetics: Systems, Ocean Engineering, IEEE Journal of Oceanic Engineering等国际学术期刊和会议中发表基金标注SCI/EI检索论文83篇，其中SCI检索期刊论文39篇（含录用）,EI检索论文44篇。论文被包括IEEE Fellow在内的多个国家和地区的学者SCI他引100余次。在多船舶编队控制领域授权国家发明专利4项。研究成果获得辽宁省科学技术二等奖1项、辽宁省自然科学学术成果奖2项、香江学者奖1项，辽宁省优秀博士论文提名奖1项。项目组成员入选国家香江学者计划（2015），辽宁省百千万人才工程千层次（2016）, 大连市科技之星（2015）等人才计划。该项目的完成从理论和实践两方面推动了多船舶编队控制研究的进步和发展。