共查询到19条相似文献,搜索用时 62 毫秒
1.
蔡玉良;吕春慧;何强;于波;陈东岳;王友童;王强;刘宇轩;赵婧婧 《计算机科学》2025,52(2):279-290
使用强化学习(Reinforcement Learning,RL)方法和基于事件驱动的完全分布式控制策略来研究系统模型信息未知的多智能体系统(MASs)的二分一致性问题。首先,基于状态阈值和时间阈值提出了一种混合事件触发机制,用于减少智能体间的通信频率。其次,利用局部采样的状态信息设计了一个自适应事件触发一致性控制协议,使所有追随智能体的二分一致性误差最终趋于零。通过排除有限时间内的芝诺(Zeno)行为,证实了上述事件触发机制的有效性。然后,基于RL方法提出了一种无模型算法来获得反馈增益矩阵,并在模型信息未知的情况下实现了自适应事件触发控制策略的构建。与现有的相关工作不同,这种基于RL的事件触发自适应控制算法只依赖于局部采样的状态信息,与任何模型信息或全局网络信息无关。此外,将上述结果扩展到切换拓扑情形,这更具挑战性,因为状态估计在以下两种情况下更新:1)交互图切换时;2)事件触发机制满足时。最后,通过实例验证了上述自适应事件触发控制算法的有效性。 相似文献
2.
喻昕;王思鉴 《小型微型计算机系统》2024,(7):1608-1614
本文提出了一种可以解决带有凸集约束且全局目标函数是非光滑伪凸的分布式优化问题的连续时间算法.本文所提出的分布式优化算法采用罚函数方法建模,通过构造一个合适的惩罚项迫使所有智能体的状态变量进入可行域.通过使用李雅普诺夫(有限时间)稳定性理论等工具,证明了在一定假设下,本文所提出的算法将在有限时间内达成一致状态且进入可行域并永驻其中,并在此之后收敛到原分布式优化问题的最优解集.此外,该算法在运行时各个智能体仅仅需要接收邻居节点的相对状态的符号信息.最后,仿真结果验证了本文所提出的算法的有效性. 相似文献
3.
4.
综述了多智能体系统分布式一致性问题的研究现状。从理论层面介绍了一致性问题的几种常见定义及与特性相关的主要参数;总结归纳了近年来几种一致性协议及其理论分析结果;分析和阐述了一致性问题的主要应用领域的进展。展望了未来的研究方向。 相似文献
5.
6.
7.
8.
9.
针对当前局部搜索算法在求解大规模、高密度的分布式约束优化问题(DCOP)时,求解困难且难以跳出局部最优取得进一步优化等问题,提出一种基于局部并行搜索的分布式约束优化算法框架(LPOS),算法中agent通过自身的取值并行地搜索局部所有邻居取值来进一步扩大对解空间的搜索,从而避免算法过早陷入局部最优。为了保证算法的收敛性与稳定性,设计了一种自适应平衡因子K来平衡算法对解的开发和继承能力,并在理论层面证明了并行搜索优化算法可以扩大对解空间的搜索,自适应平衡因子K可以实现平衡目的。综合实验结果表明,基于该算法框架的算法在求解低密度和高密度DCOP时性能都优于目前最新的算法。特别是在求解高密度DCOP中有显著的提升。 相似文献
10.
本文基于权重不平衡有向网络,对一类分布式约束优化问题进行研究,其中全局目标函数等于具有李普希兹梯度的强凸目标函数之和,并且每个智能体的状态都有一个局部约束集.每个智能体仅知道自身的局部目标函数和非空约束集.本文的目标是用分布式方法求解该问题的最优解.针对优化问题,提出了一种新的分布式投影梯度连续时间协调算法,利用拉普拉斯矩阵的零特征值对应的左特征向量消除了图的不平衡性.在某些假设下,结合凸分析理论和李雅普诺夫稳定性理论,证明了算法能够获得问题的最优解.最后,通过仿真验证了算法的有效性. 相似文献
11.
In this article, we study the finite-time consensus in probability for stochastic multi-agent systems. First, we give the nonlinear consensus protocol for multi-agent systems with Gaussian white noise, and define the concept of finite-time consensus in probability. Second, we prove that multi-agent systems can achieve the finite-time consensus in probability under five different kinds of communication topologies by using graph theory, stochastic Lyapunov theory and probability theory. Finally, some simulation examples are provided to illustrate the effectiveness of the theoretical results. 相似文献
12.
Jiayun Liu 《International journal of systems science》2016,47(16):3876-3887
This paper studies the distributed convex optimisation problem over directed networks. Motivated by practical considerations, we propose a novel distributed zero-gradient-sum optimisation algorithm with event-triggered communication. Therefore, communication and control updates just occur at discrete instants when some predefined condition satisfies. Thus, compared with the time-driven distributed optimisation algorithms, the proposed algorithm has the advantages of less energy consumption and less communication cost. Based on Lyapunov approaches, we show that the proposed algorithm makes the system states asymptotically converge to the solution of the problem exponentially fast and the Zeno behaviour is excluded. Finally, simulation example is given to illustrate the effectiveness of the proposed algorithm. 相似文献
13.
Regulatory control of distributed systems subjected to load disturbances is considered by using feedforward and state measure control configurations. Dynamic compensation of the feedforward signal is accomplished with a lead-lag function, the time constants of which are determined by means of a numerical search technique. Compensation of the state measure signal is provided by the distributed nature of the process itself. Exit temperature regulation of a tubular heat exchanger acted upon by velocity and inlet temperature disturbances is considered as an application for feedforward control. Considerably better performance is obtained with the addition of dynamic compensation to the feedforward signal. State measure control is applied to the exchanger for a feed temperature upset and the effects of sensor location on outlet performance are investigated. An optimal sensor location is determined which minimizes the integral-square error at the outlet. 相似文献
14.
针对具有符号有向图的一阶多智能体系统,研究了其固定时间二分一致性问题。为降低控制过程中多智能体系统的能量损耗,提出了一种基于事件触发机制的分布式控制协议。以图论和李雅普诺夫理论为主要研究工具,给出了多智能体系统在所设计控制协议作用下实现固定时间二分一致性的充分条件和与系统初始状态无关的收敛时间上界,并证明了使用该协议可以有效避免对智能体的连续控制和Zeno行为的发生。数值算例验证了所得理论结果的有效性。 相似文献
15.
研究了拓扑结构为有向强连通图的多智能体系统的一致性问题。提出一种新的自适应事件触发控制方案,根据采样数据动态调整触发时间间隔,提高了系统的控制性能。基于Lyapunov稳定性理论和线性矩阵不等式方法,得到多智能系统渐进趋于一致的充分条件。同时,得到一致性条件与触发参数和通信拓扑有关的结论,数值仿真结果进一步验证了理论分析的有效性。 相似文献
16.
本文研究了无向通信拓扑下二阶多智能体系统的一致性问题, 分别针对有领导者和无领导者的情形, 设计了一类基于辅助动态变量的完全分布式事件触发控制策略, 该策略具有参数较少且易调等特点. 智能体自身的触发函数满足条件时才向邻居广播自身的状态信息, 有效避免了连续通信, 减少了系统能量耗散. 每个智能体的控制协议和触发函数都只用到自身的状态和邻居触发时刻的状态, 不涉及邻居的实时状态信息, 也不依赖通信拓扑网络的任何全局信息. 利用代数图论以及Lyapunov稳定性理论, 证明在所提出的控制策略下, 二阶多智能体系统能够实现渐近一致性, 且不存在Zeno行为. 仿真示例进一步验证了理论结果的有效性. 相似文献
17.
本文研究了具有非线性不确定性的多智能体系统在切换拓扑下的预定时间一致性问题,提出了基于事件触发的分布式预定时间一致性控制算法.该算法在节约资源的情况下,可使系统在预设时间内实现一致.通过代数图论和Lyapunov理论证明系统的预定时间稳定性,且不存在Zeno行为.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,本算法的收敛时间界与初始状态无关,可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,本算法的预设收敛时间与控制算法参数无关,设定简单,且预估收敛时间具有更低的保守性. 相似文献
18.
This paper is concerned with sampled-data consensus for multi-agent systems with singular dynamics. It is assumed that the sampling period of each agent is independent of the other's. Based on event-triggered sampled-data transmission strategy, a distributed consensus protocol is presented. The consensus of singular multi-agent system is transformed into the stability of singular systems with multiple time-varying delays. By employing the Lyapunov-Krasovskii functional method, a sufficient condition on the consensus of multi-agent singular system is derived. Based on the obtained condition, an algorithm to design consensus controller gains is presented in terms of linear matrix inequalities. Two numerical examples are given to show the effectiveness of the proposed method. 相似文献
19.
Xiangkai Wu;Yuxiao Liu;Jiang Long;Wei Wang; 《国际强度与非线性控制杂志
》2024,34(8):4978-4998
》2024,34(8):4978-4998
In this article, the distributed adaptive asymptotical consensus tracking control is considered for a class of high-order nonlinear multi-agent systems with matched unknown parameters under the event-triggered communication mechanism. The controllers, adaptive laws and triggering conditions are presented under directed topology with a spanning tree, relieving communication burden among the connected agents since continuous monitoring of in-neighbors is no longer needed. To handle the cross-coupling term associated with asymmetric Laplacian matrix and unknown parameters of all agents when the candidate Lyapunov function is defined in terms of local consensus errors, adaptive immersion and invariance control technique is adopted. Adaptive estimators are also designed to estimate the constants related to global information, with which the proposed control scheme is fully distributed. It is shown that all closed-loop system signals are globally uniformly bounded and all agents can track the reference signal asymptotically while ruling out Zeno behavior though a subset of the agents cannot have direct access to the reference signal information. A numerical example is illustrated to show the effectiveness of the proposed control protocol. 相似文献