Predictive control for networked systems affected by correlated packet loss

In this paper, we consider the predictive control problem of designing receding horizon controllers for networked linear systems subject to random packet loss in the controller to actuator link. The packet dropouts are temporarily correlated in the sense that they obey a Markovian transition model. Our design task is to solve the optimal controller that minimizes a given receding horizon cost function, using the available packet loss history. Due to the correlated nature of the packet loss, standard linear quadratic regulator methods do not apply. We first present the optimal control law by considering the correlations. This controller turns out to depend on the packet loss history and would typically require a large lookup table for implementation when the Markovian order is high. To address this issue, we present and compare several suboptimal design approaches to reduce the number of control laws.     

Robust control for uncertain nonlinear systems with state‐dependent control direction

The control direction is the multiplier of the control term. In this study, the control direction is considered as an unknown function of state and allowed to cross zero and change its sign smoothly. Based on the analysis of system dynamics at the points where the control direction is zero, a robust controller is proposed by integrating with a bounding function and a Nussbaum‐type gain. Under the proposed controller, the system approaches zero if zero is accessible, or to the accessible point closest to zero if zero is not accessible by any control. The control performance is illustrated by simulated examples. Copyright © 2010 John Wiley & Sons, Ltd.     

Event‐triggered fault‐tolerant control and scheduling codesign for nonlinear networked control systems with medium‐access constraint and packet disordering

This paper studies an event‐triggered communication, scheduling, and fault‐tolerant control codesign method for nonlinear networked control systems with medium‐access constraint, delay, and packet disordering using an adaptive approximation method and adaptive technique. By considering nonlinear dynamics and controller reconfiguration, a novel event‐triggering scheme with an adjustable triggering condition and adaptive triggering thresholds is proposed. The stochastic event‐driven actuator scheduling is investigated without the assumption that the controller can access the current modes of the actuators. By considering the Markovian delay and focusing on the transmitter node, a new packet reordering approach is used to cope with packet disordering. This paper proposes an active fault‐tolerant control method, in which the nominal controller is redesigned for the postfault plant by using the fault information provided by an estimator. It is proven that the estimation error of the estimator is uniformly bounded, the reconfigurable controller and event‐trigger ensure the boundedness in probability of the state tracking error before and after the fault occurrence in the presence of medium‐access constraint, delay, and packet disordering while reducing communication load. The effectiveness of the proposed method is demonstrated in the numerical example.     

具有随机长时延的网络控制系统保性能控制

本文研究了一类具有随机长时延离散时间网络控制系统(NCSs)的建模和保性能控制问题.用两个马尔可夫链分别来描述反馈通道和前向通道的网络诱导时延,基于采用现有最新数据的原则,将闭环NCS建模成一个与当前时刻和过去时刻网络诱导时延有关的马尔可夫随机时滞系统.应用线性矩阵不等式技术和李亚普诺夫方法得到了闭环NCS随机稳定且具有二次性能指标上界的充分条件,并给出了状态反馈保性能控制器的设计方法.最后,通过数值算例验证了本文方法的有效性.     

网络控制系统的仿真研究

利用TrueTime对服从固定延时的网络控制系统进行了具体仿真,得到了较好的结果,说明TrueTime是一种较好的网络控制系统仿真工具.另外利用SimEvents对基于网络延时服从均匀分布的随机网络控制系统进行仿真,得到较好结果,说明SimEvents是一种对离散事件仿真的有效工具,同时也说明推导的随机最优控制律来控制带有随机延时所介绍的网络控制系统也是可行的.     

具有随机时延的网络控制系统周期反馈控制

针对一类具有随机时延的网络控制系统(NCSs)周期反馈控制问题,提出一个新颖的马尔可夫随机时滞系统模型.该模型包括一个分段常数广义采样保持函数(PCGSHF)以及一个用于描述网络诱导时延和数据包丢失的马尔可夫链.应用李亚普诺夫方法以及线性矩阵不等式技术得到了闭环 NCSs 随机稳定的充分条件,并给出了状态反馈控制器和 PCGSHF 的设计方法.最后,通过仿真算例验证了设计方法的有效性.     

具有数据包丢失的网络控制系统保性能控制

本文研究了一类具有数据包丢失的网络控制系统(NCSs)的建模和保性能控制问题.通过用两个马尔可夫链分别来描述前向通道和反馈通道的丢包过程,将闭环网络控制系统建模成具有两个模式的马尔可夫随机切换系统.基于线性矩阵不等式技术和李亚普诺夫方法得到了闭环系统随机稳定的充分条件,并给出了状态反馈保性能控制器的设计方法.最后通过数值算例验证本文结果的有效性.     

事件驱动的网络化系统最优控制

针对随机事件驱动的网络化控制系统, 研究其中的有限时域和无限时域内最优控制器的设计问题. 首先, 根据执行器介质访问机制将网络化控制系统建模为具有多个状态的马尔科夫跳变系统; 然后, 基于动态规划和马尔科夫跳变线性系统理论设计满足二次型性能指标的最优控制序列, 通过求解耦合黎卡提方程的镇定解, 给出最优控制律的计算方法, 使得网络化控制系统均方指数稳定; 最后, 通过仿真实验表明了所提出方法的有效性.

     

An adaptive event‐triggering scheme for networked interconnected control system with stochastic uncertainty

This paper investigates the problem of adaptive event‐triggering scheme for networked interconnected systems to relieve the burden of the network bandwidth. The data releasing is triggered by an adaptive event‐triggering device. The triggering condition depends on the state information at both the latest releasing instant and the current sampling instant. The threshold of the triggering parameter is achieved online rather than a predetermined constant. Taking the network‐induced delays and the coupling delays of the subsystems into account, together with the hybrid adaptive event‐triggering scheme and the stochastic uncertainty, we propose an unified model of the networked interconnected system. Sufficient conditions for the mean square stability and stabilization of the interconnected systems are developed by using Lyapunov–Krasovskii functional approach. A co‐designed method is put forward to obtain the controller gains and the weight of the triggering condition simultaneously. Finally, an example is provided to demonstrate the design method. Copyright © 2016 John Wiley & Sons, Ltd.     

Fault detection filter design for stochastic networked control systems

This paper is concerned with the fault detection filter (FDF) design for networked control systems subject to time‐varying transmission intervals and delays, packet dropouts, and communication constraints. The considered communication constraint is that only one network node is allowed to gain access to the shared communication channel. Also, the accessing of each node is scheduled by a specified stochastic protocol, and the remote FDFs perform the FD task only with these partially available measurements. By focus on the network‐induced phenomena, the whole FD system are first modeled in the framework of switched stochastic systems with multiple stochastic parameters. Subsequently, by using the multi‐Lyapunov functional approach and novel analysis approach, less conservative conditions including some previous existing results are derived to construct such FDFs. Finally, an example is given to illustrate the effectiveness of the proposed method.Copyright © 2013 John Wiley & Sons, Ltd.     

Modified performance of MIMO wireless networked time‐delay systems over fading channels

In this paper, the modified performance limitation of multi‐input multi‐output (MIMO) wireless networked time‐delay systems over fading channels is investigated, where the characteristic of the fading channels is assumed to be a random process. The modified tracking performance limitation is considered based on frequency domain representation. The display expressions of performance limitation are achieved by using the co‐prime factorization and the spectral decomposition techniques. The obtained results show that the modified performance limitation of the wireless networked time‐delay systems is related to the inherent characteristics of the given plant, including the non‐minimum phase zeroes, the unstable poles, the encoding‐decoding, the fading channels, the bandwidth and the modified factor. Finally, some numerical examples are provided to demonstrate the efficiency of the proposed design method.     

Observer-based controller design for networked control systems with sensor quantisation and random communication delay

This article addresses the study of observer-based controller design for network-based control systems in the presence of output quantisation and random communication delay simultaneously. In the communication channel, the output measurement are quantised before transmission, and two kinds of network-induced delays are taken into account simultaneously: (i) random delay from sensor to controller and (ii) random delay from controller to actuator. These two types of random delays are modelled as two independent Bernoulli distributed white sequences. The observer-based controller is synthesised to stabilise the networked closed-loop system in the sense of stochastic stability. Sufficient conditions for the existence of the controller are provided by stochastic Lyapunov method. An illustrative numerical example is employed to demonstrate the applicability and flexibility of the proposed design strategy.     

Small-world architecture of networked control systems

The paper proposes a method to generate communication structures for networked controllers that have the small-world property. This property, which is known from complex networks, means that in a large ring of agents, a small number of shortcut links substantially improve the performance of the overall system. To investigate the consequences of this property for control systems, the paper considers multi-agent systems with leader–follower structure where additional communication links among the local control stations are selected by the agents in a random way. The performance of the overall system is evaluated in terms of the delay with which the agents react on leader commands. The main result provides a relation between the probability p of the existence of additional edges in the communication graph and the performance of the overall system. It is proved that for any probability p > 0 there exists an upper bound of the performance index and that this bound is independent of the number of agents. Furthermore, a quantitative analysis shows that it is sufficient to connect the agents with a very low probability p ≈ 10% in order to get a satisfactory performance. These results are illustrated by a robot formation example.     

具有数据包丢失的网络控制系统的容错控制

针对一类多输入多输出网络控制系统,将数据包丢失视为系统发生执行器或传感器故障,并用最近一次的采样数据代替本周期未能成功传输的数据,对系统进行容错控制器的设计。首先以控制量丢失为例,基于Lyapunov稳定性理论,推导出了使系统稳定的条件,然后推广到检测量丢失的情况。最后通过仿真示例验证了文中方法的有效性,并对仿真结果进行了分析。     

Sampled‐data control of nonlinear networked systems with time‐delay and quantization

In this paper, the stabilization problem of a class of nonlinear networked systems with time‐delay and quantization through sampled‐data control is investigated. With sampled‐data control, sufficient conditions to guarantee the trivial solution of the nonlinear networked system to be asymptotically stable without any quantization or with uniform quantization are derived. Finally, an example of a continuous‐time nonlinear system controlled through a digital controller and a communication channel is given to illustrate the effectiveness of the proposed control methods. Copyright © 2015 John Wiley & Sons, Ltd.     

具有双边随机时延和丢包的网络控制系统有限时间控制

针对双边丢包和双边时延的网络化控制系统的有限时间控制问题,通过引入时间偏移量的方法,将丢包转换为时延,从而形成多时延系统,并将时间延迟转换为系统状态延迟。基于全概率公式给出已知网络丢包率下连续丢包的概率公式,用Markov链表征网络时延的变化规律。以线性矩阵不等式的形式给出改进的有限时间时延相关稳定性判据,并通过数值仿真验证所提方法的有效性。     

基于信噪比受限通道的网络控制系统镇定

研究了同时考虑通信和网络特性的线性系统镇定问题.首先针对连续时间网络控制系统,使用线性矩阵不等式方法得到了系统镇定所需要的最小信噪比;然后将问题推广到一类更为复杂的离散时间异构网络控制系统并得到了相似的结论.其中建立的多传感器随机时滞模型很好地刻画了token-passing总线控制网络,它更具有实际意义;最后通过数值例子对以上理论进行了仿真验证,有效表明了系统镇定时被控对象不稳定极点、网络时滞和信噪比之间的关系.     

具有均匀量化器的非线性网络控制系统的一致有界性

建立一种非线性连续被控对象、系统总时延小于一个采样周期的网络控制系统连续模型.通过线性化及离散化的方法将非线性系统转换一种离散化的模型.利用Lyapunov稳定性理论和线性矩阵不等式方法,设计控制方法使系统稳定.在系统包含均匀量化器的情况下,设计控制方法使系统一致有界.通过调节量化器的误差范围,可以将系统状态控制在一定的范围内.实际中,据此可设计相应的量化级别及编码长度等参数.通过仿真验证控制方法的有效性,并分析了量化误差对系统收敛性的影响.     

具有有限连续丢包网络控制系统的最优线性估值器

本文研究了观测数据和控制输入数据传输具有有限连续丢包的线性离散随机系统的最优估计问题.利用两个满足Bernoulli分布的随机变量来分别描述从传感器到估值器和从控制器到执行器之间的数据丢包现象.通过引入两组新的变量,将原系统转化为一个带有随机参数的系统.利用射影理论,提出了线性最小方差最优线性估值器,包括滤波器、预报器和平滑器.最后研究了稳态线性估值器,并给出了稳态存在的一个充分条件.仿真例子验证了算法的有效性.