具有随机非线性和部分转移概率未知的马尔科夫系统的H_∞控制(英文)

本文研究了非线性马尔科夫跳变系统的H∞控制问题.研究上述系统时:1)利用了非线性的概率分布信息;2)利用了转移概率中已知部分和未知部分的关系.利用李雅普诺夫泛函方法和线性矩阵不等式方法,本文得到了使得系统随机稳定的充分条件并得到了相应的反馈控制增益.文中最后给出的例子表明了所建立模型和分析方法的有效性.     

非线性时变系统部分指数稳定的一次近似

通过非线性系统的线性化方法,讨论了一类非线性时变微分系统的解关于部分变量指数稳定的一次近似.利用齐次线性系统的Cauchy矩阵解、截断Cauchy矩阵解和Gronwall_Bellman不等式,得到了线性系统的解部分指数稳定确保原非线性系统的解局部部分指数稳定的充分条件,其中一些结果可以保证部分变量有不同的指数收敛率.最后给出了一些实例说明了所用方法的有效性.     

利用稳定零点构造降阶H∞控制器的方法

现有的针对奇异情形的H∞降阶控制器的基于线性矩阵不等式(LMI)的构造算法仅利用了系统不稳定的不变零点,而没有利用系统的稳定零点.本论文试图通过对系统矩阵A引入不确定性来利用这些稳定零点,即将奇异系统矩阵A变为A0 αI(α<0),以A0代替A和系统的其它部分构成新系统,从而使得原来系统的稳定零点成为新系统的不稳定零点,进而使用降阶控制器算法得到低阶控制器.一个简单的算例表明了该方法的有效性.     

一类多变量Hammerstein模型的参数辨识方法

通过对系统输入信号的设计,使Hammerstein系统输出只反映系统的线性动态,并将非线性部分的静态影响有效地分离掉.利用最小二乘辨识得到系统的线性动态模型.基于此模型并依据系统的测量输出重构系统的中间输入,进而可估计出非线性部分的参数,据此给出了多变量Hammerstein系统辨识的动态分离方法.仿真结果表明所提出的方法是有效的.     

非线性Hammerstein系统辨识的动态分离方法

利用同幅值的M序列和逆M序列作为输入信号, 对Hammerstein模型中的线性动态部分进行分离处理, 通过辨识得到一个线性动态模型. 基于此线性模型, 依据系统的测量输出重构出系统的中间输入. 最后由系统的测试输入和中间输入估计出非线性部分的参数. 仿真结果表明本方法的有效性.     

一类不确定广义系统的分散容错控制

讨论一类不确定广义系统分散容错控制器设计问题.首先利用线性矩阵不等式(LMI)设计分散状态反馈控制器,使得广义系统执行器未出现故障时渐近稳定;接着针对广义系统的部分执行器出现故障的情况设计分散状态反馈控制器,使得闭环广义系统渐近稳定;进而利用LMI设计广义系统在分散状态反馈作用下具有完整性的容错控制器;同时对传感器故障情形设计了广义系统在分散输出反馈作用下具有完整性的容错控制器,得到了不确定广义系统关于执行器和传感器的分散容错控制器设计的方法.将所设计的控制器用于实际电子网络系统,验证了所提出方法的有效性.     

时滞切换不确定神经网络系统的指数稳定性

本文研究了具有无穷时滞切换不确定细胞神经网络(UCNNs)系统任意切换下的指数稳定性.利用同胚映射和M-矩阵理论,得到UCNNs系统平衡点存在性,唯一性和指数稳定性的充分条件;利用Lyapunov泛函方法,研究了时滞切换UCNNs系统任意切换下的鲁棒指数稳定性,并得到确保系统全局指数稳定的充分条件.     

线性不确定时滞系统的可靠保性能鲁棒控制

研究了含有故障执行器的线性不确定时滞系统的可靠保性能鲁棒控制问题.执行器故 障模型概括了执行器正常、执行器部分退化和执行器完全失效三种情况,系统的性能函数是带 有故障输入项的积分二次函数.利用Lyapunov稳定理论及矩阵不等式方法,得到了系统存在可 靠保性能鲁棒控制器的代数矩阵不等式(AMI)形式的充分条件和控制器的设计方法,给出了系 统的一个可保性能的表达式.基于LMI方法,给出了AMI的一个求解算法.一个数值例子说明 结论的可行性.     

谐振式光纤陀螺环路锁频技术研究

谐振式光纤陀螺(Resonator Fiber Optic Gyro,R-FOG)是基于Sagnac效应产生的谐振频率差来测量旋转角速率的一种新型光学传感器.对基于调相谱检测技术R-FOG系统中的环路频率锁定技术进行了研究.通过对系统光学回路和处理电路部分分别进行建模,利用反馈控制系统理论,分析了整个环路的传递函数,得到了整个环路起主导作用的简化模型.利用该简化模型,在一定的光学回路参数条件下,得到了处理电路的最佳锁定参数,并进一步在实验中得到验证.     

基于一般Lur'e系统的混沌同步研究

许多混沌系统可以转化为一个Lur'e系统.本文采用时滞反馈控制技术研究了一般 Lur'e系统的混沌同步问题,改进了最近有关文献的结果.通过利用Lyapunov-Krasovskii函数 法得到了时滞无关的线性矩阵不等式(LMI)判据,而且还通过利用M-矩阵等方法,得到了易于 检验的时滞相关与时滞无关的代数判据,利用这些结果给出了这种反馈控制器的设计方法;最 后,给出的例子阐明了所得到的结果.     

Robust stability of non-standard nonlinear singularly perturbed discrete systems with uncertainties

In the past several decades, the singularly perturbed discrete systems have received much attention for the stability analysis and controller design. Recently, there are some results about the nonlinear singularly perturbed discrete systems. Compared with the existing result, we consider the robust stability of the uncertain nonlinear singularly perturbed discrete systems with the less conservative assumption via the Lyapunov function method. Moreover, the previous results of the singularly perturbed discrete system are only applied to the system, which is composed of the slow part and the fast part, separately. However, we consider the non-standard nonlinear singularly perturbed discrete system in which the slow part and the fast part coexist, that is, a general case of the nonlinear singularly perturbed discrete systems. Then, by using the lower-order subsystems from two standard systems, we present the robust stability of the non-standard nonlinear singularly perturbed discrete system with uncertainties.     

Application of Partial-Order Methods to Reactive Programs with Event Memorization

We are concerned in this paperwith the verification of reactive systems with event memorization.The reactive systems are specified with an asynchronous reactivelanguage Electre the main feature of whichis the capability of memorizing occurrences of events in orderto process them later. This memory capability is quite interestingfor specifying reactive systems but leads to a verification modelwith a dramatically large number of states (due to the storedoccurrences of events). In this paper, we show that partial-ordermethods can be applied successfuly for verification purposeson our model of reactive programs with event memorization. Themain points of our work are two-fold: (1) we show that the independencerelation which is a key point for applying partial-order methodscan be extracted automatically from an Electreprogram; (2) the partial-order technique turns out to be veryefficient and may lead to a drastic reduction in the number ofstates of the model as demonstrated by a real-life industrialcase study.     

嵌入式实时操作系统μC/OS定时器服务的改进

实时操作系统对定时器精确度的要求大大高于通用操作系统,本文结合嵌入式实时操作系统μC/OS的具体特点,改进了它的时钟处理系统,包括细化时钟粒度、增加定时器服务。     

Cyber–physical systems technologies and application — Part II

The second part of the special issue presents more practical issues on cyber–physical systems technologies and application. We have selected five research papers whose topics are strongly related to this special issue. As continued from the part 1, we have selected 5 additional papers.     

Emergence of intelligence in next-generation manufacturing systems

In the paper we propose a fundamental shift from the present manufacturing concepts and problem solving approaches towards new manufacturing paradigms involving phenomena such as emergence, intelligence, non-determinism, complexity, self-organization, bottom-up organization, and coexistence with the ecosystem. In the first part of the paper we study the characteristics of the past and the present manufacturing concepts and the problems they caused. According to the analogy with the terms in cognitive psychology four types of problems occurring in complex manufacturing systems are identified. Then, appropriateness of various intelligent systems for solving of these four types of problems is analyzed. In the second part of the paper, we study two completely different problems. These two problems are (1) identification of system in metal forming industry and (2) autonomous robot system in manufacturing environment. A genetic-based approach that imitates integration of living cells into tissues, organs, and organisms is used. The paper clearly shows how the state of the stable global order (i.e., the intelligence) of the overall system gradually emerges as a result of low-level interactions between entities of which the system consists and the environment.     

Cooperative access systems

In the next five years we can expect to find a variety of cooperative natural language (NL) interfaces to database systems, expert systems, operating systems, CAD systems and even text formatting systems. Although the general problem of getting computers to understand NL is far from being solved, the technology for limited NL access systems is available now and there is a large growing market for such cooperative interfaces to a wide variety of software. In the first part of the paper, we illustrate the capabilities for cooperative response generation implemented in AI systems. Then we briefly review the state of the art in NL interfaces to databases and expert systems. Topics covered include user modeling, the generation of explanations, mixed initiative dialogs and knowledge acquisition via NL communication.     

Balanced truncation for unstable infinite dimensional systems via reciprocal transformation

We discuss a model reduction for unstable infinite dimensional systems using balanced truncation of reciprocal systems. The systems considered are assumed to be exponentially stabilizable and detectable, with bounded and finite rank input and output operators. We decompose the systems into their stable and unstable parts and perform reciprocal transformation only on the stable part. Balanced truncation is carried out on the reciprocal systems. This yield is then translated using reciprocal transformation as the reduced-order model of the stable part. Finally, we added the unstable part to the reduction of the stable part as the overall reduced-order model for the systems. To verify the effectiveness of the proposed approach, numerical simulations are applied to the heat equation. The performance of the proposed reduction method is compared to the balanced truncation method.     

Graph theoretic foundations of multibody dynamics

This is the first part of two papers that use concepts from graph theory to obtain a deeper understanding of the mathematical foundations of multibody dynamics. The key contribution is the development of a unifying framework that shows that key analytical results and computational algorithms in multibody dynamics are a direct consequence of structural properties and require minimal assumptions about the specific nature of the underlying multibody system. This first part focuses on identifying the abstract graph theoretic structural properties of spatial operator techniques in multibody dynamics. The second part paper exploits these structural properties to develop a broad spectrum of analytical results and computational algorithms.Towards this, we begin with the notion of graph adjacency matrices and generalize it to define block-weighted adjacency (BWA) matrices and their 1-resolvents. Previously developed spatial operators are shown to be special cases of such BWA matrices and their 1-resolvents. These properties are shown to hold broadly for serial and tree topology multibody systems. Specializations of the BWA and 1-resolvent matrices are referred to as spatial kernel operators (SKO) and spatial propagation operators (SPO). These operators and their special properties provide the foundation for the analytical and algorithmic techniques developed in the companion paper.We also use the graph theory concepts to study the topology induced sparsity structure of these operators and the system mass matrix. Similarity transformations of these operators are also studied. While the detailed development is done for the case of rigid-link multibody systems, the extension of these techniques to a broader class of systems (e.g. deformable links) are illustrated.     

Towards a learning networked organisation: human capital, compatibility and usability in e-learning systems

In all parts of organisations there flourish developments of different new subsystems in areas of knowledge and learning. Over recent decades, new systems for classification of jobs have emerged both at the level of organisations and at a macro-labour market level. Recent developments in job evaluation systems make it possible to cope with the new demands for equity at work (between, for example, genders, races, physical abilities). Other systems have emerged to describe job requirements in terms of skills, knowledge and competence. Systems for learning at work and web-based learning have created a demand for new ways to classify and to understand the process of learning. Often these new systems have been taken from other areas of the organisation not directly concerned with facilitating workplace learning. All these new systems are of course closely interrelated but, in most organisations, a major problem is the severe lack of cohesion and compatibility between the different subsystems. The aim of this paper is to propose a basis for how different human resource systems can be integrated into the business development of an organisation. We discuss this problem and develop proposals alternative to integrated macro-systems. A key element in our proposition is a structure for classification of knowledge and skill to be used in all parts of the process. This structure should be used as an added dimension or an overlay on all other subsystems of the total process. This will facilitate a continued use of all existing systems within different organisations. We develop Burge's (personal communication) model for learning to show that learning is not a successive linear process, but rather an iterative process. In this way we emphasise the need for greater involvement of learners in the development of learning systems towards increased usability in a networked system. This paper is divided into two parts which are closely related. The first part gives an overview of the lack of compatibility between the different subsystems. In this first part we note two paradoxes which impact learning and for which we propose solutions. The second part deals with 'usability' aspects of these competency-related systems; in particular, usability in e-learning systems. In this second part we describe an example of a new organisational structure. We conclude by discussing four key concepts that are necessary conditions for organisations to address when developing their human capital. Establishing these conditions helps ensure compatibility and usability in e-learning systems.     

Networks of Processes with Parameterized State Space

In general, the verification of parameterized networks is undecidable. In recent years there has been a lot of research to identify subclasses of parameterized systems for which certain properties are decidable. Some of the results are based on finite abstractions of the parameterized system in order to use model-checking techniques to establish those properties. In a previous paper we presented a method which allows to compute abstractions of a parameterized system modeled in the decidable logic WS1S. These WS1S systems provide an intuitive way to describe parameterized systems of finite state processes. In practice however, the processes in the network themselves are infinite because of unbounded data structures. One source of unboundedness can be the usage of a parameterized data structure. Another typical source may be the presence of structures ranging over subsets of participating processes. E.g., this is the case for group membership or distributed shared memory consistency protocols. In this paper we use deductive methods to deal with such networks where the data structure is parameterized by the number of processes and an extra parameter. We show how to derive an abstract WS1S system which can be subject to algorithmic verification. For illustration of the method we verify the correctness of a distributed shared memory consistency protocol using PVS for the deductive verification part and the tools PAX and SMV for the algorithmic part.