基于时变时滞动态拓扑的二阶Leader-Following多智能体一致性分析

针对具有时变时滞的二阶Leader-Following多智能体系统,研究了其一致性问题。假设其通信拓扑是时刻切换的,并且每个子时间段内系统拓扑不完全连通,采用Lyapunov-Krasovskii泛函和矛盾分析法,对系统进行了解耦分析,得出只需所有时间段内的拓扑并集连通,且系统的特征根满足一定条件,系统能达到一致,并进行了理论证明。以4个following智能体与1个leader智能体做成的网络为例进行具体说明,所得仿真结果验证了理论的有效性。     

离散时间多智能体系统一致性的平均驻留时间条件

研究高阶离散时间线性多智能体系统在有向切换信息拓扑下的状态一致性问题。首先通过提出的线性变换将该一致性问题转换为相应离散时间线性切换系统的渐近稳定性问题。然后借助于切换系统稳定性的平均驻留时间方法,分别得到如下两种情形下该一致性问题可解的充分条件：1)信息拓扑集合中的一部分拓扑是可一致的;2)信息拓扑集合中所有信息拓扑是可一致的。最后通过数值仿真验证了所得理论结果的正确性。     

Edge event-triggered control for multi-agent systems under directed communication topologies

The edge event-triggered consensus control for multi-agent systems with single- or double-integrator dynamics under directed communication topologies is investigated in this paper. With the edge event-triggered sampling strategies, a class of novel consensus protocols is proposed to reduce control update frequency. The main features of the proposed edge event-triggered control include three aspects: (i) measurement errors are defined based on directed information flow and thus they allow that the communication topology is directed, (ii) each individual controller updates its output only when the associated edge event occurs and the control actions are independent of other agents, and (iii) the proposed edge event-triggered control does not exhibit Zeno behaviour and it is shown that the inter-event intervals are lower bounded by a positive constant. In the consensus analysis, matrix transformation techniques are used to convert the high-dimensional system into a low-dimensional system, and some sufficient consensus conditions are derived for the multi-agent systems with single- or double-integrator dynamics. Finally, the effectiveness of the edge event-triggered sampling strategies is illustrated by simulations.     

高阶有向积分器网络的包含控制

研究高阶固定有向积分器网络的包含控制问题。针对静止和运动的领导者,分别为跟随者提出了不同的控制协议。领导者是静止的情况下,应用拉普拉斯变换终值定理提出线性控制协议,驱使跟随者渐近进入由多个领导者所构成的静态凸包中;领导者是移动的情况下,提出非线性控制协议,以保证跟随者能够渐近进入由多个领导者所构成的动态凸包中,并使跟随者一直保持在该凸包中,同时提出了能够达到这一目标的充分条件。仿真结果验证了该理论分析的正确性。     

联合连通条件下的二阶多智能体系统有限时间一致性控制

针对时变动态拓扑下无leader的二阶多智能体系统有限时间一致性控制问题,本文给出有限时间一致性协议,并对所提出的一致性协议进行理论分析.基于图论、Lyapunov稳定性理论、同次性理论和积分不等式方法,证明了当通信拓扑为联合连通时,有限时间一致性协议可保证系统在有限时间达到一致.最后给出仿真结果,验证了理论的有效性.     

时延多智能体系统领导跟随一致性研究

为符合实际情形,针对不确定与随机发生非线性的多智能体系统,研究了有时延且网络拓扑切换时系统的领导跟随一致性。传统协议通常保守的假设邻接个体间通信时延与个体和领导者间通信时延大小相同,新协议中上述时延可以大小不同。相比于传统方法,新颖的将复杂网络同步问题研究领域对时延的处理方法引入到多智能体系统一致性问题的研究中,利用一类推广的Halanay不等式,给出系统实现领导跟随一致性需满足的两个与时延无关的充分条件,即时延在相关参数满足定理条件的前提下不影响系统最终实现一致性。相比其他方法得到的含有时延的判定条件,本研究结果保守性更低,实例仿真验证了新协议的可行性。     

Leader-following consensus of fractional-order multi-agent systems via adaptive pinning control

In this paper, the leader-following consensus problem of fractional-order multi-agent systems is considered via adaptive pinning control. The dynamics of leader and all followers with linear and nonlinear functions are investigated, respectively. We assume that the node should be pinned if its in-degree is less than its out-degree in the paper. Under this assumption and based on the stability theory of fractional-order differential systems, some leader-following consensus criteria are derived, which are easily obtained by matrix inequalities. The control of each agent using local information is designed and detailed analysis of the leader-following consensus is presented. The design technique is based on algebraic graph theory and the Riccati inequality. Several simulation examples are presented to demonstrate the effectiveness of the proposed method.     

Leader-following consensus of multiple uncertain Euler–Lagrange systems under switching network topology

In recent years, the leader-following consensus problem for multiple uncertain Euler–Lagrange systems has been studied under some restrictive assumptions on the network topology. In this paper, we further study the same problem under switching network topology. We propose a distributed adaptive control law that can solve the problem under a switching network satisfying jointly connected condition. Under this condition, our results do not require the network to be undirected and allow the network to be disconnected at any time instant. Moreover, by introducing an exosystem to generate various reference signals, our control law can handle a class of reference signals such as sinusoidal signals with arbitrary amplitudes and initial phases or ramp signals with arbitrary slopes.     

非线性随机多智能体系统的固定时间一致性

本文研究了在固定拓扑和切换拓扑下,非线性随机多智能体系统的固定时间一致性问题.首先针对固定拓扑,设计了一种非线性控制协议,利用随机Lyapunov稳定性理论和代数图论给出了实现固定时间一致性的充分条件和收敛时间的上界值,随后将结论推广至切换拓扑,设计的切换拓扑子图的并集只需要满足连通条件,即可实现固定时间一致,模型更具...     

Controllability of multi-agent systems with periodically switching topologies and switching leaders

This paper considers controllability of multi-agent systems with periodically switching topologies and switching leaders. The concept of m-periodic controllability is proposed, and a criterion for m-periodic controllability is established. The effect of the duration of subsystems on controllability is analysed by utilising a property of analytic functions. In addition, the influence of switching periods on controllability is investigated, and an algorithm is proposed to search for the fewest periods to ensure controllability. A necessary condition for m-periodic controllability is obtained from the perspective of eigenvectors of the subsystems’ Laplacian matrices. For a system with switching leaders, it is proved that switching-leader controllability is equivalent to multiple-leader controllability. Furthermore, both the switching order and the tenure of agents being leaders have no effect on the controllability. Some examples are provided to illustrate the theoretical results.     

具有切换拓扑的多智能体系统协同控制的差分隐私保护

网络信息技术的不断发展与普及使得各类数据的发布采集变得方便与便捷, 但数据的直接发布势必会造成个网络信息的泄露和敏感信息的失密, 因此敏感信息的保护成为了各行各业关注的问题. 本文研究了基于固定拓扑和切换拓扑的多智能体系统协同控制的差分隐私保护问题, 将差分隐私算法与传统平均一致性算法结合, 提出了具有隐私保护的协同控制算法, 分析了隐私保护算法对分布式协同控制闭环系统稳定性的影响. 基于所提算法, 应用矩阵论和概率统计对隐私保护协同控制算法的收敛性和隐私性进行理论分析, 该算法可以保护智能个体的数据隐私, 同时可以使得系统运动实现均方一致. 在系统拓扑结构动态变化的情况下, 本文对该算法的收敛性和隐私性进行理论分析, 讨论了切换拓扑对隐私保护的影响. 最后的仿真示例验证了理论结果的正确性.