Showing papers in "Studies in Informatics and Control in 2012"

Artificial Bee Colony (ABC) Algorithm for Constrained Optimization Improved with Genetic Operators

Abstract: Artificial bee colony (ABC) is a relatively new swarm intelligence based metaheuristic. It was successfully applied to unconstrained optimization problems and later it was adjusted for constrained problems as well. In this paper we introduce modifications to the ABC algorithm for constrained optimization problems that improve performance of the algorithm. Modifications are based on genetic algorithm (GA) operators and are applied to the creation of new candidate solutions. We implemented our modified algorithm and tested it on 13 standard benchmark functions. The results were compared to the results of the latest (2011) Karaboga and Akay’s ABC algorithm and other state-of-the-art algorithms where our modified algorithm showed improved performance considering best solutions and even more considering

Influence of Specific AR Capabilities on the Learning Effectiveness and Efficiency

Abstract: Augmented reality (AR) is a promising technology for improving the application and comprehension skills of students. The ARiSE project developed an Augmented Reality Teaching Platform (ARTP) for secondary schools. A Chemistry learning scenario was implemented that is based on the interaction paradigm “building with guidance”. This study aims at assessing the extent to which specific capabilities of the ARTP are supporting the understanding Chemistry concepts as well as their contribution to the perceived utility. The results of a multiple-regression analysis shows that the specific features of the Chemistry scenario are supporting a better understanding and learning with less effort. Overall, the interaction paradigm proved to have a positive influence on the effectiveness and efficiency of the learning process.

Physical Programming: A Review of the State of the Art

Abstract: Most traditional multi-criteria optimization techniques require that the decision maker construct an aggregate objective function using the weights determined as a result of a trial and error process. Physical programming (PP) eliminates this tedious weight assignment process by providing decision makers with a flexible and more natural problem formulation. In PP, the decision maker specifies ranges of different degrees of desirability instead of defining weights. In this paper, we present a comprehensive review of PP studies by classifying them into four major categories (viz., methodological papers, industrial engineering applications, mechanical engineering applications and other applications). Insights from the review and future research directions conclude the paper.

Taylor and Bi-local Piecewise Approximations with Neuro-Fuzzy Systems

Abstract: Methods of local approximations based on truncated Taylor series are well understood; several methods to compute bounds for the approximation errors are available (Christensen and Christensen, 2006), (Powell, 1981), (Shahriari, 2006). The use of Taylor approximations is convenient, among others, because methods to determine the approximation accuracy are well established. Taylor approximations are widely used in solving differential equations (Kloeden, Platen, 1995), (Tachev, 2009), control applications (Hedjar et al., 2005), circuit modeling (Jridi and Alfalou, 2009), data processing, image generation, signal prediction (Hedjar et al., 2005), and economic modeling (Judd, 1998).

A Hybrid Particle Swarm Optimization – Simulated Annealing Algorithm for the Probabilistic Travelling Salesman Problem

Abstract: The Probabilistic Traveling Salesman Problem (PTSP) is a variation of the well known Traveling Salesman Problem (TSP). This problem arises when the information about customers demand is not available at the moment of the tour generation and/or the tour re-calculating cost is too elevated. In this article, a Hybrid Algorithm combining Particle Swarm Optimization (PSO) and Simulated Annealing (SA) is proposed, in order to solve the PTSP. The PSO heuristic offers a simple structured algorithm which supplies a high level of exploration and fast convergence, compared with other evolutionary algorithms. The SA algorithm is used to improve the particle diversity and to avoid the algorithm being trapped into local optimum. Two well-known benchmarks of the literature are used and the proposed PSO-SA algorithm obtains acceptable results. In fact, the hybrid algorithm improves the performance of simple PSO algorithm for all instances.

Decentralized Formation Control of Multi-agent Robot Systems based on Formation Graphs

Abstract: Formation control is an important issue of motion coordination of Multi-agent Robots Systems. The goal is to coordinate a group of agents to achieve a desired formation pattern. The control strategies are decentralized because every robot does not possess information about the positions and goals of all the other robots. Based on the formation graphs properties and the local potential functions approach, we obtain a formal result about global convergence to the desired pattern for any formation graph. Also, we characterize the topologies of the formation graphs where the centroid of positions remains stationary. Finally, the control approach is extended to the case of unicycle-type robots.

Simulation of Adaptive LQ Control of Nonlinear Process

Abstract: The use of the computer simulation not only in the control engineering grows rapidly nowadays with the increasing speed of the computers and low prices of the hardware. Furthermore, the simulation is very often used at present as it has many advantages over an experiment on a real system, which is not feasible and can be dangerous, time and money demanding. A modelling of the system usually precedes the simulation [1]. The mathematical model is a kind of abstract representation of the process which uses input, state or output variables, relations between these variables collected in the set of mathematical equations [1] and [2]. Some simulation and modelling examples can be found also in [3] and [4].

Optimization of Robotic Mobile Agent Navigation

Abstract: The traveling salesman problem (TSP) has many applications in economy, transport logic [1] etc. It also has a wide range of applicability in the mobile robot path planning optimization [2]. The paper presents research result of solving the path planning subproblem of the navigation of an intelligent autonomous mobile robotic agent. Collecting objects by a mobile robotic agent is the final problem that is intended to be solved. For the robotic mobile agent’s path planning is used an unsupervised neural network that can find a closely optimal path between two points in the agent’s working area. We have considered a modification of the criteria function of the winner neuron selection. Simulation results are discussed at the end of the paper. The next future development is the hardware implementation of the selforganizing map with real time functioning.

Performance/Robustness Trade-off Design Framework for 2DoF PI Controllers

Abstract: The aim of the paper is to present a design framework for two-degree-of-freedom (2DoF) proportional integral (PI) controllers that allows to deal with the control system performance/robustness trade-off. It is based on the use of a model reference optimization procedure with target servo-control and regulatory control closed-loop transfer functions for firstand second-order-plus-dead-time (FOPDT, SOPDT) models. A smooth servo/regulatory combined performance is obtained by forcing both closed-loop transfer functions to perform as close as possible to target non-oscillatory dynamics. A comparison with other methods shows the effectiveness of the proposed design methodology.

A Decision Support Tool Based on a Portfolio Selection Model for Crop Planning under Risk

Abstract: In this paper is presented a decision support tool for crop planning under risk. The software tool is based on a portfolio selection model for crop planning. The portfolio selection model is a minimum financial risk model. It takes into account climate risk and market risk. The decision variable describes the land allocation to crops. The model was solved with the MINLP solver from GAMS. The decision support tool has an interface that facilitates the construction of the input data collection and the user parameters. It gives a flexible way of working and is mainly user oriented. Numerical results obtained with this Decision Support tool are analyzed.

Multimodel control design using unsupervised classifiers

Abstract: Multimodel approaches derive a smooth control law from the blending of local controllers using the concept of validities and domain overlapping. In this paper, it is demonstrated that unsupervised classification algorithms can be of a great help to design such parameters as the number of the models and their respective clusters, which will be performed using a respectively Rival Penalized Competitive Learning (RPCL) and simple or fuzzy K-means algorithms. The classical multimodel approach follows by deriving parametric model identification using the classification results for models orders and then parameters estimation. The determination of the global system control parameters results from a fusion of models control parameters. The case of a second order nonlinear system is studied to illustrate the efficiency of the proposed approach, and it is shown that this approach is much simpler that other multimodel control design methods which generally require a huge number of neighboring models.

An Integrated Feature Selection and Classification Scheme

Abstract: Irrelevant and redundant features may not only deteriorate the performances of classifiers, but also slow the prediction process. Another problem in prediction is the availability of a large number of classification models. How to choose a satisfactory classifier is an important yet understudied task. The goal of this paper is to propose an integrated scheme for feature selection and classifier evaluation in the context of prediction. It combines traditional feature selection techniques and multi-criteria decision making (MCDM) methods in an attempt to increase the accuracies of classification models and identify appropriate classifiers for different types of data sets.

Feedback Linearization and Model Reference Adaptive Control of a Magnetic Levitation System

Abstract: In recent years, due to computational developments that have enabled more complex applications of nonlinear problems, the area of nonlinear control systems has been the subject of many studies (Soltanpour and Shafie, 2010). The present paper shows a combination of an adaptive controller and a feedback linearization technique to control a Magnetic Levitation System (MLS). This system was chosen since it has nonlinear dynamics and a didactic kit of the physical system is available to continue with future work.

Analyzing the Performance of Spectrum Sensing Algorithms for IEEE 802.11 af Standard in Cognitive Radio Network

Abstract: In the entire world the wireless communication systems are represented by 2G and 3G systems with all stages of evolution towards 4G systems. The complexity of wireless networks requires a careful design, especially related to bandwidth and energy efficiency. Bandwidth efficiency is very important parameter, because it relates to frequency spectrum, which is a natural limited resource. The cognitive radio has been proposed as the future technology to meet the ever increasing demand of the radio spectrum by allocating the spectrum dynamically to allow unlicensed access on noninterfering basis. The digital dividend of 700MHz band (mainly used by TV broadcast services) opens the door for cognitive radio applications due to its excellent propagation characteristics compared to GSM 1800 MHz, 2.1 GHz or 2.5 GHz bands. In cognitive radio, spectrum sensing is the fundamental problem. In this paper we are analyzing the performance of spectrum sensing algorithms; energy detection and covariance absolute value utilizing TV white space for IEEE 802.11 af standard.

Using Software Metrics for Automatic Software Design Improvement

Abstract: In this paper we are approaching the problem of improving the quality of a software system design, an important issue during the evolution of object oriented software systems. Starting from the fact that software metrics are essential in measuring the software quality, we introduce a metric based high dimensional representation of the elements of a software system (application classes and methods from the application classes) and we define a distance semi-metric between the elements of the software system. An experimental validation of the distance semi-metric on two case studies is provided. The obtained results illustrate that the distance function introduced in this paper can be successfully used for improving the internal structure of software systems, highlighting this way the potential of our proposal.

An Easy to Apply Methodology for Fault Detection and Isolation in Linear Systems

Abstract: Although the fault detection and isolation problem has been there for over three decades and many solutions using various approaches has been proposed, simpler and feasible methods to quickly and efficiently solve this problem are still valuable. In this paper, a very easy to implement method to solve the fault detection and isolation problem for linear systems is proposed. Experimental result using the three-tank system, commonly used as a benchmark, shows the feasibility of the proposed method. The technique is based on an unusual application of a long standing result of the geometric control theory for linear systems, initially proposed for the disturbance decoupling problem. Nevertheless it avoids the involved calculations that are usual in geometric-based algorithms.

Bringing EU Entrepreneurs together through Cross-border Services SPOCS - a Case Study

Abstract: The work presented in this paper highlights the continuous improvement that has been made in terms of economic competitiveness of the European companies, with an emphasis on Small and Medium - sized Enterprises (SME's) by enabling all businesses - national and European - to benefit from available, efficient and interoperable electronic procedures.

On-line Master/Slave Robot System Synchronization with Obstacle Avoidance

Abstract: In this work, it is proposed a controller for the synchronization of master/slave robotic systems. The aim of the proposed controller is to provide autonomy to the slave robot, via obstacle avoidance capability. The controller includes two terms. The first term is a PID controller, which is mapped through the task Jacobian from the task space to the robot joint space. The second term is the on-line solution of an optimal control problem (OCP), which considers the dynamic model of the slave robot. The performance index of the OCP pursues three objectives. The first goal is synchronization of master/slave end-effector position. The second goal is to keep the joint positions of the slave robot within feasible limits. The third goal is the obstacle avoidance of the whole arm at the slave robot side. Experimental results show the effectiveness of our proposal when tested on a writing task.

Syntax extensions for a constrained-object language via dynamic parser cooperation

Abstract: A modern feature of constraint languages is the ability of compiling a model into a set of solver languages. This allows one to model a problem in a single language and to execute it in a set of solver engines. The idea is to facilitate experimentation as well as model sharing. The common architecture to support this task is composed of three layers: an upper layer for the modeling language, a bottom layer for the solver language, and a middle one for performing the mapping process. However, this architecture has an important inconvenience: there is no mechanism for updating the modeling language. This paper addresses this concern by introducing a simple description language for extending the syntax of the modeling language. The goal is to make the architecture adaptable to further upgrades of the solver layer.

Usability Evaluation of Clinician Web Back-Ends to Telemonitoring Systems: Two Case-Studies in Scotland

Abstract: One of the reasons why large scale deployment of telemedicine has not been successful is the difficulty healthcare workers have in managing the software. Good usability is essential to the success of a telemedicine solution. By ensuring that user needs are efficiently and effectively respected, usability encourages user acceptance and reduces the need for support. However, little is known about what healthcare workers require from telemedicine systems in terms of how patient acquired data is displayed and interrogated. This paper describes the results of a post-implementation investigation of the clinician web ‘back-ends’ of two telemedicine systems used for the monitoring of long-term conditions in Lothian, Scotland, focusing on the features healthcare workers would like to see in future systems. We conducted semi-structured interviews and questionnaires to ascertain the views of healthcare workers who had been using the systems. The results of the evaluation were used to design a new prototype generic telemonitoring website which we offered to participants to demonstrate possible improvements and to further seek their views. The prototype was very well received, participants considering that it was easier to use and more user friendly than the system that they had been using.

Classifier Evaluation for Software Defect Prediction

Abstract: Feature selection is an essential step in the process of software defect prediction due to the negative effect of irrelevant features on classification algorithms. Hence selecting the most relevant and representative features is critical to the success of software defect detection. Another problem in software defect prediction is the availability of a large number of classification models. This paper applies feature selection and classifier evaluation in the context of software defect prediction. An empirical study is presented to validate the proposed scheme using 9 classifiers over 4 public domain software defect data sets. The results indicate that the proposed scheme can improve the performance of classifiers using the most representative features and recommend classifiers that are accurate and reliable in software defect prediction.

Formal Information Representation for Tactical Reconnaissance System Organization Model

Abstract: The purpose of this paper is to explore formal information representation for tactical reconnaissance system (TRS) organization model, i.e. to describe its framework and analyze its description logic to form organization information in establishing a logical simulation model for the real organization system. This paper establishes system caste diagram, analyzes task, role and entity for TRS organization, and uses Object-Z specification language to design relevant formal models of organization, roles and entities in TRS. It further presents a method in transforming Object-Z class tree to TRS task tree organization, and checks these formal models by implementing simulation. This work proposes a novel approach of formal information representation for a complex warfare system organization model through transforming subsystem-level representation to Object-Z representation, till task tree that reflects system organization information. The application in tactical reconnaissance task decomposition modelling and simulation demonstration system proves that this approach is feasible and effective.

Arrangement of Face-to-Face Meetings using Social Media

Abstract: Social networks, especially Facebook as currently the largest one, made the organization of events apparently simpler. Facebook offers the event service, which has greatly simplified the invitation process. Still, organizing an event is usually coupled with the risk of guest's no-show. We report an investigation designed to identify factors that might help predict a person's likelihood of attendance to an event s/he is invited to. Our research tries to combine information research with information technology tool design. The factors affecting the probability were determined by an analysis of data acquired in surveys among hundred and fifty or so Facebook users. We also developed a program that implements (some of) these findings. Simple quantitative and qualitative analyses were carried out on the data, sufficient to identify some of the key factors influencing invitees in their decision to attend a meeting. The factors as identified by the surveys are indeed relevant to attendance of meetings arranged using social media. Our application can help the event creator estimate how many people would attend his event and predict the likelihood of each invitee's attendance. Moreover, the application can also help the invited guest learn which of his friends are likely to attend the same event. Research that combines information side with information technology side can be fruitful as shown by this simple result. There is room for future work in this interdisciplinary space.

Control of a Modular Multilevel Matrix Converter for High Power Applications

Abstract: A control scheme for a multilevel AC-AC converter is presented. The converter resembles a matrix converter but it uses the cascade converter in place of converter valve. This is a string of H-bridge modules, each equipped with a DC storage capacitor, as the building block of the converter. This yields a highly modular implementation approach which may be suitable for high voltage, high power applications. One of the main issues with these modular multilevel converter topologies is the regulation of the capacitors voltages. This is a prerequisite to successfully operate the topology. This work explains the basics of the multilevel matrix converter and develops its associated control schemes to operate it as AC-AC converter. Control objectives include: regulation of multiple capacitor voltages; control of the multiple valve currents; and control of the output voltages. The proposed scheme is verified experimentally and through simulations. Results show good system performance in rejecting load impacts; maintaining capacitor voltages and valve currents within normal values; drawing low distortion, and close to unitary power factor, currents from the line; and creating high quality output voltages which result in low distortion load currents. The matrix multilevel converter thus shows great potential for high performance, high power motor drives.

Distributed Neural Networks Microcontroller Implementation and Applications

Abstract: Abstract: In this paper it is argued that, for any three-layer perceptron, it is always possible to design an equivalent distributed ANN, wherein the neurons are implemented on the nodes of a communication network, and the synapses between them are established in the communication process. In this approach, neurons are seen as processing and communication entities. Since both local and distributed implementations of a specific ANN are perfectly equivalent, they can use the same set of synapse weights, i.e. a distributed ANN can be trained on a local, equivalent software implementation. Two use cases are presented to demonstrate the validity of the idea.

Movement and Color Detection of a Dynamic Object: An application to a Mobile Robot

Abstract: The paper describes the integration of several image processing algorithms necessary to recognize a particular color and the movement of an object. The main objective is to detect the object by its color and track it by a mobile robot. Mean filter is applied to soften and sharpen the input image. Then, RGB filter is applied to calculate the center of mass and area of the object and to locate its position in a real environment to develop the robot motion. These algorithms are applied to a mobile robot, in a tested scenario, tracking an object.

Evolutionary Method for Designing and Learning Control Structure of a Wheelchair

Abstract: This article describes an aspect of evolutionary robotics for trajectory tracking. We will combine genetic algorithms with neural networks for modelling and controlling a wheelchair for disabled people. The interest of the hybridization of Neural Networks (NN) with Evolutionary Algorithms (EA) in robotics is based on the observation that a local search by a gradient descent method is replaced by a global search performed by EA. The gradient descent methods are subject to variations in performance due to the initial position of the NN, which sometimes leads to a convergence towards local minima. In contrast, the proposed evolutionary methods provide a global research of both the structure and the weights of the neural net. The control structure used for robot trajectory tracking control is based on the Internal Model Control (IMC) which direct neural model was learned with our new EA.

Study of Financial Systems Volatility Using Suboptimal Estimation Algorithms

Abstract: This paper presents and implements a novel stochastic volatility (SV) model, based on the structure of the GARCH model, to describe the relationship between an observed financial return series and its standard deviation, namely volatility. The proposed approach has been compared to the standard GARCH as the underlying modeling structure within a particle-filtering-based scheme for state estimation. The proposed structure has been implemented to estimate the volatility of both a simulated return series and the NASDAQ Composite index during the period July 21, 2008 July 17, 2009. The results of this procedure show that the parameters of the GARCH model can be used in the uGARCH structure, allowing the latter representing the hidden volatility as accurate as the standard GARCH, but also providing an estimate of the whole probability density.