切换拓扑结构下多智能体系统的一致性研究

针对一类多智能体系统,研究了系统在切换拓扑结构下的一致性控制问题。假定系统拓扑结构的变化符合马尔可夫过程,设计与切换拓扑结构相关的一致性协议,利用线性矩阵不等式和随机稳定性理论,给出了多智能体系统满足一致性的充分条件,并通过数值仿真验证了所提控制方案的可行性和有效性。     

具有时延和切换拓扑的高阶离散时间多智能体系统鲁棒保性能一致性

研究存在参数不确定性的高阶离散时间多智能体系统在时延和联合连通切换通信拓扑条件下的鲁棒保性能一致性问题,给出一种线性一致性协议的设计方法.1）引入高阶离散时间不确定多智能体系统的鲁棒保性能一致性问题,定义基于智能体邻居状态误差和控制输入的保性能函数;2）通过构造合适的Lyapunov函数并利用离散时间系统稳定性理论,推导出一个使高阶离散时间不确定多智能体系统在该条件下获得保性能一致性的线性矩阵不等式（Linear matrix inequality,LMI）充分条件,并给出相应的保性能上界;3）以一致性序列的形式给出参数不确定条件下的高阶离散时间多智能体系统的一致性收敛结果;4）数值仿真验证了本文理论的正确性和有效性.     

离散时间混合多智能体的拟平均一致性控制

为了研究混合一阶和二阶异质无向多智能体网络离散时间一致性问题,提出了固定和可变拓扑结构的多智能体网络的拟平均一致性协议.根据Lyapunov函数和代数图论,分析了多智能体网络的稳定性,表明在固定和可变拓扑结构的情况下,混合阶多智能体系统可以达到拟平均一致.对6个节点组成的4种不同的无向网络拓扑结构进行了仿真,结果显示了变拓扑离散时间异质多智能体网络的位置和速度均能达到拟平均一致.仿真结果表明,该算法对混合阶多智能体网络的拟平均一致性控制有效.     

非平衡拓扑结构的多智能体网络系统一致性协议

针对多智能体网络系统,本文分别研究了非平衡拓扑结构的多智能体连续状态和离散状态下的一致性协议.首先提出了能使用有向图表示的多智能体网络系统的拓扑结构,并根据该拓扑结构建立了网络系统的1阶数学模型和提出了多智能体网络系统一致收敛准则.对于多智能体网络连续系统,该系统的一致平衡点最终收敛于初始状态的凸组合,本文最终确定了非平衡拓扑结构的一致平衡点.如果多智能体网络系统的拓扑结构没有改变,在离散状态下系统的一致平衡点仍收敛于初始状态的凸组合,并且离散状态下系统的一致平衡点与连续状态下系统的一致平衡点相等.最后采用8个智能体组成的网络拓扑进行计算机仿真,验证理论的正确性.     

具有时滞的混合阶多智能体系统的组一致性

针对由离散时间一阶和二阶智能体组成的混合阶多智能体系统,研究其在固定和切换拓扑结构下受通信时滞影响时的组一致性问题。分别为两类智能体提出组一致性协议,引入模型变换,将闭环系统转化为等价系统。在一定假设条件下,以代数图论、矩阵理论为主要研究工具,分别在固定和切换拓扑结构下给出了混合阶多智能体系统实现渐近组一致性的条件。采用数值仿真对所得结果的有效性进行了验证。     

一类异构多智能体系统有向图下的广义平均一致性分析

本文对有向图下离散时间一阶二阶混合的异构多智能体系统广义平均一致性问题进行研究.首先给出了该异构系统广义平均一致性的基本概念.在此基础上,针对平均一致性研究中对交互拓扑为平衡网络的局限,提出了一种基于辅助变量的线性一致性协议,对每一个智能体增加一个辅助变量,用于记录个体的状态更新.采用图论、非负矩阵理论、特征值扰动等方法进行分析证明,表明该协议使得异构多智能体系统在任意强连通有向图下达到广义平均一致性.并对收敛值的性质进行了分析.最后,通过仿真对该结论进行了验证.     

离散异质多智能体系统的分组一致性控制

针对一类混合异质多智能体系统的分组一致性控制问题进行了研究。具体分析了由一阶智能体和二阶智能体组成的混合异质系统,研究其在离散情况下的分组一致性。基于两个合理的假设提出了线性控制协议,运用代数图论、稳定性理论和矩阵理论,分析协议作用下闭环系统的系统矩阵及动态特性,取得了系统渐近实现分组一致性的充分条件,该条件与系统拓扑结构、采样周期以及控制参数有关。结论同时适用于有向拓扑与无向拓扑,最后通过仿真实例对所得分析结果进行了验证。     

高阶时滞多智能体系统一致性控制

本文讨论了多输入网络化时滞多智能体系统的一致性问题.应用无向或有向图描述网络化系统的拓扑结构.多智能体的一致性收敛问题可以转化为线性时滞系统的稳定性问题.通过双线性矩阵不等式可以得到高阶多智能体系统时滞相依以及时滞独立的一致性准则.最后,举例一个无向拓扑结构的时滞多智能体来说明本文方法的有效性.     

基于事件触发机制的多智能体网络平均一致性研究

文章研究了事件触发机制下具有固定和切换拓扑结构的多智能体网络的平均一致性问题。为了有效降低多智能体一致性控制协议的更新次数,文章提出了仅依赖于各智能体及其邻居节点信息的分布式事件触发机制。为了使多智能体网络渐近收敛至初始平均状态,文章提出了基于事件触发机制的多智能体平均一致性协议。然后,文章建立了事件触发机制下的闭环系统模型,并分别获得了固定和切换拓扑结构下多智能体网络平均一致收敛的充分条件。仿真表明,基于事件触发机制的平均一致性协议保证了多智能体网络的平均一致性,并且分布式事件触发机制有效降低了多智能体一致性协议更新次数。     

基于部分稳定性方法的离散时间多智能体系统的一致性

研究有向信息拓扑下离散时间线性多智能体系统的一致性分析与设计问题.利用提出的线性变换,将一致性问题转换为相应线性系统的部分变元渐近稳定性问题.基于部分变元稳定性理论,得到有向信息拓扑下离散时间线性多智能体系统达到渐近一致的基于矩阵Schur稳定性的充要条件和状态一致函数的解析表达式.同时设计了反馈增益矩阵.最后数值实例验证了所得理论的有效性.     

Finite-time consensus for multi-agent networks with unknown inherent nonlinear dynamics

The objective of this paper is to analyze the finite-time convergence of a nonlinear but continuous consensus algorithm for multi-agent networks with unknown inherent nonlinear dynamics. Due to the existence of the unknown inherent nonlinear dynamics, the stability analysis and the finite-time convergence analysis are more challenging than those under the well-studied consensus algorithms for known linear systems. For this purpose, we propose a novel comparison based tool. By using this tool, it is shown that the proposed nonlinear consensus algorithm can guarantee finite-time convergence if the directed switching interaction graph has a directed spanning tree at each time interval. Specifically, the finite-time convergence is shown by comparing the closed-loop system under the proposed consensus algorithm with some well-designed closed-loop system whose stability properties are easier to obtain. Moreover, the stability and the finite-time convergence of the closed-loop system using the proposed consensus algorithm under a (general) directed switching interaction graph can even be guaranteed by the stability and the finite-time convergence of some well-designed nonlinear closed-loop system under some special directed switching interaction graph. This provides a stimulating example for the potential applications of the proposed comparison based tool in the stability analysis of linear/nonlinear closed-loop systems by making use of known results in linear/nonlinear systems.     

采用超节点协同的多智能体系统一致性算法

针对如何提高多智能体系统达到一致性的收敛速度的问题,提出了一种采用超节点协同的多智能体系统一致性算法.新算法对多智能体系统建立图信号模型,在图中选出超节点进行协同,有效提高一致性收敛速度.首先利用单跳采样算法对图进行超节点的选取和局部集的划分,并对局部集内的节点进行一次协同.然后超节点之间进行边的连接得到粗化图,用粗化图的拉普拉斯矩阵特征值设计图滤波器的系数.最后超节点的信号经过图滤波器迭代达到平均值后,传输给其一阶邻居节点,使所有节点达到平均一致.仿真结果表明所提算法能够最终实现平均一致性,与现有方法相比,可以显著提高收敛速度,并减少计算量.     

一种HRG模型初始化算法及在链路预测中的应用

在利用层次随机图（HRG）模型对真实网络进行链路预测的过程中,需要构造一个初始层次随机图来初始化马尔科夫链以运行马尔科夫链蒙特卡洛抽样算法。针对现有的层次随机图初始化方案效率不高的问题,本文对初始层次随机图模型进行重建,提出一种新的层次随机图模型初始化算法。该算法分为2个阶段,第一阶段引入相似性指标（LHN-I指标）为网络中的边进行排序;第二阶段利用排序好的边对层次随机图模型进行构造。在该过程中,设计一种将网络顶点插入到层次随机图模型中的方法。通过3个实例网络对提出的算法与现有算法的性能进行比较,实验结果表明,利用提出的初始化算法构造出的初始层次随机图不仅有着较高的似然值,而且使得马尔科夫链蒙特卡洛算法能够更快地收敛,进而降低链路预测的时间消耗。除此之外,在链路预测实验中,改进的基于层次随机图模型的链路预测算法相比一些基于相似性指标的链路预测算法有着较好的预测精度。     

Combination framework of rendezvous algorithm for multi‐agent systems with limited sensing ranges

In this paper, we present a novel rendezvous algorithm framework ‐ the combination framework, which is named after the term convex combination in computational geometry. The convergence of the proposed algorithms is proved based on the novel tools from the recent results in graph theory and consensus study. It has been observed that the circumcenter algorithm can be considered as a special case of the combination framework. Moreover, the relevant analysis provides us a novel viewpoint on designing and the rendezvous rate of rendezvous algorithms for networked multi‐agent systems with limited sensing ranges. Copyright © 2010 John Wiley and Sons Asia Pte Ltd and Chinese Automatic Control Society     

人工蜂群算法的收敛性分析

利用随机过程理论,对人工蜂群算法收敛性进行理论分析,给出人工蜂群算法的一些数学定义和蜜源位置的一步转移概率,建立人工蜂群算法的Markov链模型,分析此Markov链的一些性质,论证了人工蜂群状态序列是有限齐次Markov链,且状态空间是不可约的。结合随机搜索算法的全局收敛准则,证明了人工蜂群算法能够满足随机搜索算法全局收敛的两个假设,保证算法的全局收敛。     

A general graph-based semi-supervised learning with novel class discovery

In this paper, we propose a general graph-based semi-supervised learning algorithm. The core idea of our algorithm is to not only achieve the goal of semi-supervised learning, but also to discover the latent novel class in the data, which may be unlabeled by the user. Based on the normalized weights evaluated on data graph, our algorithm is able to output the probabilities of data points belonging to the labeled classes or the novel class. We also give the theoretical interpretations for the algorithm from three viewpoints on graph, i.e., regularization framework, label propagation, and Markov random walks. Experiments on toy examples and several benchmark datasets illustrate the effectiveness of our algorithm.     

全局孤立性和局部同质性图表示的随机游走显著目标检测算法

目前的显著性检测算法主要依赖像素间的相互对比,缺乏对显著目标自身特性的分析理解. 依据显著目标是显眼、紧凑和完整的思路,提出一种基于目标全局孤立性和局部同质性的 随机游走显著目标检测算法,将视觉显著性检测公式化为马尔科夫随机游走问题. 首先将输入图像进行分块,根据像素块之间颜色特征和方向特征的相似性确定边的权重, 从而构建图模型;然后通过全连通图搜索提取全局特性,突出全局较孤立的区域; 同时通过k-regular图搜索提取局部特性,增强局部较均匀的区域;最后将全局特性和局部 特性相结合得到显著图,进而确定感兴趣区域位置. 实验结果表明,相比于其他两种具有代表性的算法,所提方法检测结果更加准确、合理, 证明该算法切实可行.     

Distributed Event‐Triggered Consensus for Multiple Underactuated Systems Under Markovian Switching Topologies

This paper presents a distributed consensus algorithm that employs event‐triggered communication for multiple underactuated systems under Markovian switching topologies. Instead of the general stochastic topology, the graph of the entire system is governed by a set of Markov chains to the edges, which can recover the general Markovian switching topologies in line with the practical communication network. By utilizing integral sliding mode control strategy, rigorous analysis of the asymptotic convergence results has been performed through graph theory and Lyapunov stability theory. An event‐triggered communication law is provided for each agent and Zeno behavior of triggering time sequences is excluded. It will yield to the very first application of the multiple underactuated systems, in which the system states could be enforced to track the leader. Finally, the illustrative simulations on six underactuated two‐link manipulators are given to demonstrate the effectiveness of theoretical results.     

Convergence Rate Analysis of Gaussian Belief Propagation for Markov Networks

Gaussian belief propagation algorithm (GaBP) is one of the most important distributed algorithms in signal processing and statistical learning involving Markov networks. It is well known that the algorithm correctly computes marginal density functions from a high dimensional joint density function over a Markov network in a finite number of iterations when the underlying Gaussian graph is acyclic. It is also known more recently that the algorithm produces correct marginal means asymptotically for cyclic Gaussian graphs under the condition of walk summability (or generalised diagonal dominance). This paper extends this convergence result further by showing that the convergence is exponential under the generalised diagonal dominance condition, and provides a simple bound for the convergence rate. Our results are derived by combining the known walk summability approach for asymptotic convergence analysis with the control systems approach for stability analysis.      

符号图下含扰动的多智能体系统预定时间二分一致性

研究符号图下具有扰动的多智能体系统二分一致性控制问题.考虑了线性系统、非线性不确定系统以及切换拓扑的情况,分别提出相应的预定时间控制器,各控制器可使系统在预定时间内实现二分一致性.通过Lyapunov稳定性理论、代数图论和矩阵分析等验证算法的准确性;仿真对比实验验证了所提算法的可行性和有效性;相较于有限时间控制算法,所提算法的收敛时间不依赖于初始状态,可以通过选择单一时间参数设定系统收敛时间上界;相较于固定时间控制算法,所提算法预设收敛时间与控制算法参数无关,设定简单,具有更低的保守性.