基于事件触发的离散多智能体二分一致性

针对一阶离散多智能体系统,研究了事件触发控制下的二分一致性问题.首先考虑智能体间通信拓扑结构为无向连通结构平衡图的情形,针对各智能体设计事件触发控制,包括仅依赖于自身及邻居智能体采样状态的控制输入,以及仅依赖自身状态的事件触发条件,实现了对通信资源的节约利用.基于图论、离散系统稳定性理论,证明系统能够实现二分一致性.同时,合理设置控制输入及事件触发条件中参数,保证系统不存在Zeno现象.之后,进一步分析设计了包含有向生成树的结构平衡图下,多智能体系统的事件触发控制.最后利用仿真实例验证了理论结果的有效性.     

基于Q-learning的离散时间多智能体系统一致性

针对模型未知的一类离散时间多智能体系统,本文提出了一种Q-learning方法实现多智能体系统的一致性控制.该方法不依赖于系统模型,能够利用系统数据迭代求解出可使给定目标函数最小的控制律,使所有智能体的状态实现一致.通过各个智能体所产生的系统数据,采用策略迭代的方法实时更新求解得到多智能体系统的控制律,并对所提Q-le...     

Consensus seeking in a network of discrete-time linear agents with communication noises

This paper studies the mean square consensus of discrete-time linear time-invariant multi-agent systems with communication noises. A distributed consensus protocol, which is composed of the agent's own state feedback and the relative states between the agent and its neighbours, is proposed. A time-varying consensus gain a[k] is applied to attenuate the effect of noises which inherits in the inaccurate measurement of relative states with neighbours. A polynomial, namely ‘parameter polynomial’, is constructed. And its coefficients are the parameters in the feedback gain vector of the proposed protocol. It turns out that the parameter polynomial plays an important role in guaranteeing the consensus of linear multi-agent systems. By the proposed protocol, necessary and sufficient conditions for mean square consensus are presented under different topology conditions: (1) if the communication topology graph has a spanning tree and every node in the graph has at least one parent node, then the mean square consensus can be achieved if and only if ∑^∞_{k = 0}a[k] = ∞, ∑^∞_{k = 0}a²[k] < ∞ and all roots of the parameter polynomial are in the unit circle; (2) if the communication topology graph has a spanning tree and there exits one node without any parent node (the leader–follower case), then the mean square consensus can be achieved if and only if ∑^∞_{k = 0}a[k] = ∞, lim_{k → ∞}a[k] = 0 and all roots of the parameter polynomial are in the unit circle; (3) if the communication topology graph does not have a spanning tree, then the mean square consensus can never be achieved. Finally, one simulation example on the multiple aircrafts system is provided to validate the theoretical analysis.     

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.     

离散时间高阶不确定线性多个体系统保性能一致性分析

本文研究了存在参数不确定性的离散时间高阶多个体系统保性能一致性问题,给出了一种设计其线性一致性协议的方法.首先,通过模型转换的方法将该问题转换为一组离散时间不确定系统的稳定性问题;然后,构造合适的Lyapunov函数并利用离散时间系统稳定性理论,推导出一个使离散时间高阶不确定多个体系统获得保性能一致的LMI充分条件;接着,以一致性序列的形式给出参数不确定条件下的离散时间高阶多个体系统的一致性收敛结果.最后,参数不确定的对比数值仿真验证了本文理论的正确性和有效性.     

An output formation consensus control for leader–follower networked multi-agent systems under communication constraints and switching topologies

In this paper, the formation consensus problem for a class of leader–follower networked multi-agent systems under communication constraints and switching topologies is investigated. A networked predictive control scheme is proposed to achieve stability and output formation consensus with the switching topology, capable of compensating for data loss and time delays in the network. By equating the whole closed-loop networked multi-agent system with the proposed control scheme to the corresponding switched system, the sufficient and necessary condition of output formation consensus and stability for agents is given. Finally, using three-degree-of-freedom air-bearing spacecraft simulators as the control objects, the proposed scheme is demonstrated to be able to actively compensate for the communication constraints through numerical simulations, and it is also verified to have a good control performance by further realizing the formation task of the simulators through practical experiments.     

Observer-based tracking controller design for networked predictive control systems with uncertain Markov delays

This paper is concerned with a tracking controller design problem for discrete-time networked predictive control systems. The control law used here is a combined state-feedback control and integral control. Since not all the states are available in practice, a local Luenberger observer is utilised to estimate the state vector. The measured output and estimated state vector are packed together and transmitted to the tracking controller via a communication channel with a limited capacity. Meanwhile, the control signal is also transmitted through a communication network.Network-induced delays on both links are considered for the signal transmission and modelled by Markov chains. Moreover, it is assumed that the elements in Markov transition matrices are subject to uncertainties. In order to fully compensate for network-induced delays, the controller generates a sequence of control signals which are dependent on each possible delay in the feedforward channel. By taking the augmentation twice, we obtain delay-free stochastic closed-loop systems and the controlled output is chosen as the tracking error. Sufficient conditions are provided for the energy-to-peak performance of the closed-loop systems. The feedback gains of the controller can be derived by solving a minimisation problem. Two examples are illustrated to demonstrate the effectiveness of the proposed design method.     

离散线性切换系统的一致有限时间稳定分析和反馈控制及其在网络控制系统中的应用

研究了一类离散线性切换系统的一致有限时间稳定性分析和反馈镇定．基于线性矩阵不等式技术,给出了在任意切换信号作用下,离散线性切换系统有限时间稳定和有限时间有界的充分条件,并给出了离散线性切换控制系统一致有限时间状态反馈控制器的设计方法．将上述分析结果应用于一类丢包有界的网络控制系统,得到了保证其有限时间稳定的反馈控制器．最后,通过两个数值仿真例子验证了所提方法的有效性．     

高阶时滞多智能体系统一致性控制(英文)

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.     

区间时滞相关离散非线性系统的鲁棒模型预测控制

针对一类带有扰动、输入约束和凸多面体不确定性的区间时滞离散非线性系统, 提出一种鲁棒模型预测控制方法. 一方面, 利用min-max 模型预测控制求解鲁棒模型预测控制器, 以研究鲁棒预测控制在范数有界意义下的扰动抑制问题; 另一方面, 充分利用时滞的上下界信息构造Lyapunov 函数以得到控制器存在的充分条件. 最后给出了闭环系统鲁棒稳定性证明.     

基于观测器的线性时变时滞多智能体系统一致性

针对系统状态不可测和具有通信时延的线性多智能体系统,提出一种基于观测器的一致性控制算法.设计观测器用于解决智能体状态不可测的问题,在观测器的基础上,提出一种控制协议来实现带时变时延的线性多智能体系统一致性.利用模型转换的方式将原系统转换为新的模型系统,在此基础上,构造Lyapunov-Krasovskii函数并分析系统稳定性,通过求解线性矩阵不等式获取控制器系数矩阵.最后通过Matlab数值仿真验证所提出方法的有效性.