Showing papers in "International Journal of Applied Mathematics and Computer Science in 2005"

Observer design for systems with unknown inputs

Abstract: Design procedures are proposed for two different classes of observers for systems withunknown inputs In thefirst approach, the state of the observed system is decomposed into known and unknown components The unknown component is a projection, not necessarily orthogonal, of the whole state along the subspace in which the available state component resides Then, a dynamical system to estimate the unknown component is constructed Combining the output of the dynamical system, which estimates the unknown state component, with the available state information results in an observer that estimates the whole state It is shown that some previously proposed observer architectures can be obtained using the projection operator approach presented in this paper The second approach combines sliding modes and the second method of Lyapunov resulting in a nonlinear observer The nonlinear component of the sliding mode observer forces the observation error into the sliding mode along a manifold in the observation error space Design algorithms are given for both types of observers

Image retrieval based on hierarchical Gabor filters

Abstract: Content Based Image Retrieval (CBIR) is now a widely investigated issue that aims at allowing users of multimedia information systems to automatically retrieve images coherent with a sample image. A way to achieve this goal is the computation of image features such as the color, texture, shape, and position of objects within images, and the use of those features as query terms. We propose to use Gabor filtration properties in order to find such appropriate features. The article presents multichannel Gabor filtering and a hierarchical image representation. Then a salient (characteristic) point detection algorithm is presented so that texture parameters are computed only in a neighborhood of salient points. We use Gabor texture features as image content descriptors and efficiently emply them to retrieve images.

Correcting spelling errors by modelling their causes

Abstract: This paper accounts for a new technique of correcting isolated words in typed texts. A language-dependent set of string substitutions reflects the surface form of errors that result from vocabulary incompetence, misspellings, or mistypings. Candidate corrections are formed by applying the substitutions to text words absent from the computer lexicon. A minimal acyclic deterministic finite automaton storing the lexicon allows quick rejection of nonsense corrections, while costs associated with the substitutions serve to rank the remaining ones. A comparison of the correction lists generated by several spellcheckers for two corpora of English spelling errors shows that our technique suggests the right words more accurately than the others.

Fast leak detection and location of gas pipelines based on an adaptive particle filter

Abstract: Leak detection and location play an important role in the management of a pipeline system. Some model-based methods, such as those based on the extended Kalman filter (EKF) or based on the strong tracking filter (STF), have been presented to solve this problem. But these methods need the nonlinear pipeline model to be linearized. Unfortunately, linearized transformations are only reliable if error propagation can be well approximated by a linear function, and this condition does not hold for a gas pipeline model. This will deteriorate the speed and accuracy of the detection and location. Particle filters are sequential Monte Carlo methods based on point mass (or “particle”) representations of probability densities, which can be applied to estimate states in nonlinear and non-Gaussian systems without linearization. Parameter estimation methods are widely used in fault detection and diagnosis (FDD), and have been applied to pipeline leak detection and location. However, the standard particle filter algorithm is not applicable to time-varying parameter estimation. To solve this problem, artificial noise has to be added to the parameters, but its variance is difficult to determine. In this paper, we propose an adaptive particle filter algorithm, in which the variance of the artificial noise can be adjusted adaptively. This method is applied to leak detection and location of gas pipelines. Simulation results show that fast and accurate leak detection and location can be achieved using this improved particle filter.

Finite horizon nonlinear predictive control by the taylor approximation: application to robot tracking trajectory

Abstract: In industrial control systems, practical interest is driven by the fact that today’s processes need to be operated under tighter performance specifications. Often these demands can only be met when process nonlinearities are explicitly considered in the controller. Nonlinear predictive control, the extension of well-established linear predictive control to nonlinear systems, appears to be a well-suited approach for this kind of problems. In this paper, an optimal nonlinear predictive control structure, which provides asymptotic tracking of smooth reference trajectories, is presented. The controller is based on a finite–horizon continuous time minimization of nonlinear predicted tracking errors. A key feature of the control law is that its implementation does not need to perform on-line optimization, and asymptotic tracking of smooth reference signal is guaranteed. An integral action is used to increase the robustness of the closed-loop system with respect to uncertainties and parameters variations. The proposed control scheme is first applied to planning motions problem of a mobile robot and, afterwards, to the trajectory tracking problem of a rigid link manipulator. Simulation results are performed to validate the tracking performance of the proposed controller.

Identifiability and estimation of pharmacokinetic parameters for the ligands of the macrophage mannose receptor

Abstract: NATHALIE VERDIERE∗, LILIANNE DENIS-VIDAL∗∗ GHISLAINE JOLY-BLANCHARD∗, DOMINIQUE DOMURADO∗∗∗ ∗ Department Genie Informatique, University of Technology of Compiegne BP 20 529, 60 205 Compiegne Cedex, France e-mail: nathalie.verdiere@dma.utc.fr, ghislaine.joly-blanchard@utc.fr ∗∗ University of Sciences and Tech. Lille, France e-mail: denvid@attglobal.net ∗∗∗ CRBA — Faculty of Pharmacy BP 14 491, 34093 Montpellier Cedex 5, France e-mail: domurado@univ-montp1.fr

A new definition of the fuzzy set

Abstract: The present fuzzy arithmetic based on Zadeh’s possibilistic extension principle and on the classic definition of a fuzzy set has many essential drawbacks. Therefore its application to the solution of practical tasks is limited. In the paper a new definition of the fuzzy set is presented. The definition allows for a considerable fuzziness decrease in the number of arithmetic operations in comparison with the results produced by the present fuzzy arithmetic.

Picture languages in automatic radiological palm interpretation

Abstract: The paper presents a new technique for cognitive analysis and recognition of pathological wrist bone lesions. This method uses AI techniques and mathematical linguistics allowing us to automatically evaluate the structure of the said bones, based on palm radiological images. Possibilities of computer interpretation of selected images, based on the methodology of automatic medical image understanding, as introduced by the authors, were created owing to the introduction of an original relational description of individual palm bones. This description was built with the use of graph linguistic formalisms already applied in artificial intelligence. The research described in this paper demonstrates that for the needs of palm bone diagnostics, specialist linguistic tools such as expansive graph grammars and EDT-label graphs are particularly well suited. Defining a graph image language adjusted to the specific features of the scientific problem described here permitted a semantic description of correct palm bone structures. It also enabled the interpretation of images showing some in-born lesions, such as additional bones or acquired lesions such as their incorrect junctions resulting from injuries and synostoses.

Design of mealy finite-state machines with the transformation of object codes

Abstract: An optimization method of the logic circuit of a Mealy finite-state machine is proposed. It is based on the transformation of object codes. The objects of the Mealy FSM are internal states and sets of microoperations. The main idea is to express the states as some functions of sets of microoperations (internal states) and tags. The application of this method is connected with the use of a special code converter in the logic circuit of an FSM. An example of application is given. The effectiveness of the proposed method is also studied.

Comparison of linear control methods for an amb system

Abstract: Department of Control, AGH University of Science and TechnologyAl. Mickiewicza 30, 30–095 Cracow, Polande-mail: {wgr,ap}@ia.agh.edu.plThe contactless nature of active magnetic bearings brings about many advantages over the conventional bearing whileindustrial real-time applications are often limited by the significant complexity of control algorithms. This paper presentsthe application of an LQ controller to an active magnetic bearing system (AMB). Two control strategies are presented andcompared: local and global. In the first case the rotor is modelled as two separated masses located at the bearing. In thesecond case rotor stabilization is considered globally as a problem of the rotating rigid body suspended in a magnetic field.The second approach is especially important for high-speed rotating machines. The control performance of both algorithmswas analysed using an experimental AMB laboratory system.

The ud rls algorithm for training feedforward neural networks

Abstract: A new algorithm for training feedforward multilayer neural networks is proposed. It is based on recursive least squares procedures and U-D factorization, which is a well-known technique in filter theory. It will be shown that due to the U-D factorization method, our algorithm requires fewer computations than the classical RLS applied to feedforward multilayer neural network training.

Range identification for a perspective dynamic system with a single homogeneous observation

Abstract: LILI MA∗, YANGQUAN CHEN∗, KEVIN L. MOORE∗∗ ∗ Center for Self-Organizing and Intelligent Systems (CSOIS) Department of Electrical and Computer Engineering, 4160 Old Main Hill Utah State University (USU), Logan, UT 84322–4160, USA e-mail: lilima@cc.usu.edu, yqchen@ece.usu.edu ∗∗ Research and Technology Development Center Johns Hopkins University Applied Physics Laboratory M/S 2–236 11100, Johns Hopkins Road Laurel, MD 20723-6099, USA e-mail: kevin.moore@jhuapl.edu

On naive Bayes in speech recognition

Abstract: The currently dominant speech recognition technology, hidden Markov modeling, has long been criticized for its simplistic assumptions about speech, and especially for the naive Bayes combination rule inherent in it. Many sophisticated alternative models have been suggested over the last decade. These, however, have demonstrated only modest improvements and brought no paradigm shift in technology. The goal of this paper is to examine why HMM performs so well in spite of its incorrect bias due to the naive Bayes assumption. To do this we create an algorithmic framework that allows us to experiment with alternative combination schemes and helps us understand the factors that influence recognition performance. From the findings we argue that the bias peculiar to the naive Bayes rule is not really detrimental to phoneme classification performance. Furthermore, it ensures consistent behavior in outlier modeling, allowing efficient management of insertion and deletion errors.

Realization problem for a class of positive continuous-time systems with delays

Abstract: In positive systems, inputs, state variables and outputs take only non-negative values. Examples of positive systems are industrial processes involving chemical reactors, heat exchangers and distillation columns, storage systems, compartmental systems, or water and atmospheric pollution models. A variety of models revealing the behaviour of positive linear systems can be found in engineering, management science, economics, social sciences, biology and medicine, etc.

Non-quadratic performance design for takagi-sugeno fuzzy systems

Abstract: This paper improves controller synthesis of discrete Takagi-Sugeno fuzzy systems based on non-quadratic Lyapunov functions, making it possible to accomplish various kinds of control performance specifications such as decay rate conditions, requirements on control input and output and disturbance rejection. These extensions can be implemented via linear matrix inequalities, which are numerically solvable with commercially available software. The controller design is illustrated with an example.

Latent Semantic Indexing for patent documents

Abstract: Since the huge database of patent documents is continuously increasing, the issue of classifying, updating and retrieving patent documents turned into an acute necessity. Therefore, we investigate the efficiency of applying Latent Semantic Indexing, an automatic indexing method of information retrieval, to some classes of patent documents from the United States Patent Classification System. We present some experiments that provide the optimal number of dimensions for the Latent Semantic Space and we compare the performance of Latent Semantic Indexing (LSI) to the Vector Space Model (VSM) technique applied to real life text documents, namely, patent documents. However, we do not strongly recommend the LSI as an improved alternative method to the VSM, since the results are not significantly better.

Optimal random sampling for spectrum estimation in dasp applications

Abstract: In this paper we analyse a class of DASP (Digital Alias-free Signal Processing) methods for spectrum estimation of sampled signals. These methods consist in sampling the processed signals at randomly selected time instants. We construct estimators of Fourier transforms of the analysed signals. The estimators are unbiased inside arbitrarily wide frequency ranges, regardless of how sparsely the signal samples are collected. In order to facilitate quality assessment of the estimators, we calculate their standard deviations. The optimal sampling scheme that minimises the variance of the resulting estimator is derived. The further analysis presented in this paper shows how sampling instant jitter deteriorates the quality of spectrum estimation. A couple of numerical examples illustrate the main thesis of the paper.

Neuro-fuzzy modelling based on a deterministic annealing approach

Abstract: This paper introduces a new learning algorithm for artificial neural networks, based on a fuzzy inference system ANBLIR. It is a computationally effective neuro-fuzzy system with parametrized fuzzy sets in the consequent parts of fuzzy if-then rules, which uses a conjunctive as well as a logical interpretation of those rules. In the original approach, the estimation of unknown system parameters was made by means of a combination of both gradient and least-squares methods. The novelty of the learning algorithm consists in the application of a deterministic annealing optimization method. It leads to an improvement in the neuro-fuzzy modelling performance. To show the validity of the introduced method, two examples of application concerning chaotic time series prediction and system identification problems are provided.

Ant algorithm for flow assignment in connection-oriented networks

Abstract: This work introduces ANB (Ant Algorithm for Non-Bifurcated Flows), a novel approach to capacitated static optimization of flows in connection-oriented computer networks. The problem considered arises naturally from several optimization problems that have recently received significant attention. The proposed ANB is an ant algorithm motivated by recent works on the application of the ant algorithm to solving various problems related to computer networks. However, few works concern the use of ant algorithms in the assignment of static flows in connection-oriented networks. We analyze the major characteristics of the ANB and try to explain its performance. We report results of many experiments over various networks.

An observability problem for a class of uncertain-parameter linear dynamic systems

Abstract: An observability problem for a class of linear, uncertain-parameter, time-invariant dynamic SISO systems is discussed. The class of systems under consideration is described by a finite dimensional state-space equation with an interval diagonal state matrix, known control and output matrices and a two-dimensional uncertain parameter space. For the system considered a simple geometric interpretation of the system spectrum can be given. The geometric interpretation of the system spectrum is the base for defining observability and non-observability areas for the discussed system. The duality principle allows us to test observablity using controllability criteria. For the uncertain-parameter system considered, some controllability criteria presented in the author’s previous papers are used. The results are illustrated with numerical examples.

Delay-dependent asymptotic stabilitzation for uncertain time-delay systems with saturating actuators

Abstract: This paper concerns the issue of robust asymptotic stabilization for uncertain time-delay systems with saturating actuators. Delay-dependent criteria for robust stabilization via linear memoryless state feedback have been obtained. The resulting upper bound on the delay time is given in terms of the solution to a Riccati equation subject to model transformation. Finally, examples are presented to show the effectiveness of our result.

Control of a team of mobile robots based on non-cooperative equilibria with partial coordination

Abstract: In this work we present an application of the concept of non-cooperative game equilibria to the design of a collision free movement of a team of mobile robots in a dynamic environment. We propose the solution to the problem of feasible control synthesis, based on a partially centralized sensory system. The control strategy based on the concept of non-cooperative game equilibria is well known in the literature. It is highly efficient through phases where the solution is unique. However, even in simple navigation problems, it happens that multiple equilibria occur, which incurs a problem for control synthesis and may lead to erroneous results. In this paper we present a solution to this problem based on the partial centralization idea. The coordinator module is incorporated into the system and becomes active when multiple equilibria are detected. The coordination method includes a “fair arbiter” for the selection of an appropriate equilibrium solution. Simulation studies of the proposed methodology were carried out for 2, 3 and 5 robots, and their results are presented.

Extension of The Cayley-Hamilton Theorem to Continuous-time Linear Systems With Delays

Abstract: The classical Cayley-Hamilton theorem (Gantmacher, 1974; Lancaster, 1969) says that every square matrix satisfies its own characteristic equation. The Cayley-Hamilton theorem was extended to rectangular matrices (Kaczorek, 1995c), block matrices (Kaczorek, 1995b; Victoria, 1982), pairs of commuting matrices (Chang and Chan, 1992; Lewis, 1982; 1986; Mertizios and Christodoulous, 1986), pairs of block matrices (Kaczorek, 1998), and standard and singular two-dimensional linear (2-D) systems (Kaczorek, 1992/93, 1994; 1995a; Smart and Barnett, 1989; Theodoru, 1989).

Monomial subdigraphs of reachable and controllable positive discrete-time systems

Abstract: A generic structure of reachable and controllable positive linear systems is given in terms of some characteristic components (monomial subdigraphs) of the digraph of a non-negative a pair. The properties of monomial subdigraphs are examined and used to derive reachability and controllability criteria in a digraph form for the general case when the system matrix A may contain zero columns. The graph-theoretic nature of these criteria makes them computationally more efficient than their known equivalents. The criteria identify not only the reachability and controllability properties of positive linear systems, but also their reachable and controllable parts (subsystems) when the system does not possess such properties.

Hybrid stabilization of discrete-time LTI systems with two quantized signals

Abstract: GUISHENG ZHAI∗, YUUKI MATSUMOTO∗∗, XINKAI CHEN∗∗∗ JOE IMAE∗, TOMOAKI KOBAYASHI∗ ∗ Department of Mechanical Engineering Osaka Prefecture University Sakai, Osaka 599–8531, Japan e-mail: zhai@me.osakafu-u.ac.jp ∗∗ Department of Opto-Mechatronics Wakayama University Sakaedani, Wakayama 640–8510, Japan ∗∗∗ Department of Electronic and Information Systems Shibaura Institute of Technology Saitama 337–8570, Japan e-mail: chen@sic.shibaura-it.ac.jp

Colored decision process Petri nets: modeling, analysis and stability

Abstract: In this paper we introduce a new modeling paradigm for developing a decision process representation called the Colored Decision Process Petri Net (CDPPN). It extends the Colored Petri Net (CPN) theoretic approach including Markov decision processes. CPNs are used for process representation taking advantage of the formal semantic and the graphical display. A Markov decision process is utilized as a tool for trajectory planning via a utility function. The main point of the CDPPN is its ability to represent the mark-dynamic and trajectory-dynamic properties of a decision process. Within the mark-dynamic properties framework we show that CDPPN theoretic notions of equilibrium and stability are those of the CPN. In the trajectory-dynamic properties framework, we optimize the utility function used for trajectory planning in the CDPPN by a Lyapunov-like function, obtaining as a result new characterizations for final decision points (optimum point) and stability. Moreover, we show that CDPPN mark-dynamic and Lyapunov trajectory-dynamic properties of equilibrium, stability and final decision points converge under certain restrictions. We propose an algorithm for optimum trajectory planning that makes use of the graphical representation (CPN) and the utility function. Moreover, we consider some results and discuss possible directions for further research.

Image recall using a large scale generalized brain-state-in-a-box neural network

Abstract: An image recall system using a large scale associative memory employing the generalized Brain-State-in-a-Box (gBSB) neural network model is proposed. The gBSB neural network can store binary vectors as stable equilibrium points. This property is used to store images in the gBSB memory. When a noisy image is presented as an input to the gBSB network, the gBSB net processes it to filter out the noise. The overlapping decomposition method is utilized to efficiently process images using their binary representation. Furthermore, the uniform quantization is employed to reduce the size of the data representation of the images. Simulation results for monochrome gray scale and color images are presented. Also, a hybrid gBSB-McCulloch-Pitts neural model is introduced and an image recall system is built around this neural net. Simulation results for this model are presented and compared with the results for the system employing the gBSB neural model.

Uncertainty Models of Vision Sensors in Mobile Robot Positioning

Abstract: This paper discusses how uncertainty models of vision-based positioning sensors can be used to support the planning and optimization of positioning actions for mobile robots. Two sensor types are considered: a global vision with overhead cameras, and an on-board camera observing artificial landmarks. The developed sensor models are applied to optimize robot positioning actions in a distributed system of mobile robots and monitoring sensors, and to plan the sequence of actions for a robot cooperating with the external infrastructure supporting its navigation.

On the stability of neutral-type uncertain systems with multiple time delays

Abstract: The problems of both single and multiple delays for neutral-type uncertain systems are considered. First, for single neutral time delay systems, based on a Razumikhin-type theorem, some delay-dependent stability criteria are derived in terms of the Lyapunov equation for various classes of model transformation and decomposition techniques. Second, robust control for neutral systems with multiple time delays is considered. Finally, we demonstrate numerical examples to illustrate the effectiveness of the proposed approaches. Compared with results existing in the literature, our methods are shown to be superior to them.

A novel continuous model to approximate time Petri nets: modelling and analysis

Abstract: In order to approximate discrete-event systems in which there exist considerable states and events, David and Alla define a continuous Petri net (CPN). So far, CPNs have been a useful tool not only for approximating discrete-event systems but also for modelling continuous processes. Due to different ways of calculating instantaneous firing speeds of transitions, various continuous Petri net models, such as the CCPN (constant speed CPN), VCPN (variable speed CPN) and the ACPN (asymptotic CPN), have been proposed, where the continuous flow is specified uniquely by maximal firing speeds. However, in applications such as chemical processes there exist situations where the continuous flow must be above some minimal speed or in the range of minimal and maximal speeds. In this paper, from the point of view of approximating a time Petri net, the CPN is augmented with maximal and minimal firing speeds, and a novel continuous model, i.e., the Interval speed CPN (ICPN) is defined. The enabling and firing semantics of transitions of the ICPN are discussed, and the facilitating of continuous transitions is classified into three levels: 0-level, 1-level and 2-level. Some policies to resolve the conflicts and algorithms to undertake the behavioural analysis for the ICPN are developed. In addition, a chemical process example is presented.