当前位置: X-MOL 学术IEEE Trans. Cybern. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Nonfragile Near-Optimal Control of Stochastic Time-Varying Multiagent Systems With Control- and State-Dependent Noises
IEEE Transactions on Cybernetics ( IF 11.8 ) Pub Date : 2019-07-01 , DOI: 10.1109/tcyb.2018.2829713
Yuan Yuan , Zidong Wang , Peng Zhang , Hongli Dong

In this paper, the near-optimal nonfragile consensus control design problem is investigated for a class of discrete time-varying multiagent systems (MASs) with control- and state-dependent noises. A decentralized observer-based control protocol is proposed by using the relative output measurements. The gain perturbations/variations of the controller as well as the state- and control-dependent noises are simultaneously taken into consideration, which could better reflect the complexities in reality. The corresponding time-varying observer-based nonfragile near-optimal consensus protocol is designed for the underlying MASs over a finite horizon. To be specific, a certain upper bound is first derived for the associate cost function for the MASs. Then, such an upper bound is minimized by using the completing-the-square technique and Moore–Penrose pseudo inverse. The parameters of the time-varying observer/controller are obtained in terms of the solutions to the Riccati-like recursion. In virtue of the matrix partitioning technique, the explicit expressions of the control/observer parameters are presented. Finally, based on the derived consensus protocol, an upper bound of the associate cost function is provided as time goes to infinity. Some numerical simulations are conducted to demonstrate the validity of the proposed methodology.

中文翻译:

具有控制和状态相关噪声的随机时变多智能体系统的非脆弱近最优控制

本文研究了一类具有控制和状态相关噪声的离散时变多主体系统(MAS)的近似最优非脆弱共识控制设计问题。通过使用相对输出测量值,提出了一种基于分散观察者的控制协议。同时考虑了控制器的增益扰动/变化以及与状态和控制相关的噪声,这可以更好地反映现实中的复杂性。相应的基于时变观察者的非脆弱近最优共识协议是为有限范围内的基础MAS设计的。具体而言,首先为MAS的关联成本函数导出某个上限。然后,通过使用完全平方技术和Moore-Penrose伪逆,可以最大程度地减小这样的上限。时变观察者/控制器的参数是根据类似Riccati递归的解获得的。借助于矩阵划分技术,给出了控制/观察者参数的显式表达式。最后,基于派生的共识协议,随着时间的流逝,无穷大提供了关联成本函数的上限。进行了一些数值模拟,以证明所提出方法的有效性。基于派生的共识协议,随着时间的流逝,关联成本函数的上限被提供。进行了一些数值模拟,以证明所提出方法的有效性。基于派生的共识协议,随着时间的流逝,关联成本函数的上限被提供。进行了一些数值模拟,以证明所提出方法的有效性。
更新日期:2019-07-01
down
wechat
bug