Eugen IANCU*, Ionela IANCU**
*Department of Automation, Electronics and Mechatronics, University of Craiova, 107 Decebal Street, RO-200440 Craiova, Romania (e-mail: Eugen.Iancu@automation.ucv.ro, http://www.ace.ucv.ro)
**Department of Physiology, University of Medicine and Pharmacy of Craiova, 4, Petru Rares Street, RO-200349 Craiova, Romania (e-mail: iancu_ep@yahoo.com, http://www.umfcv.ro)

Abstract: Diabetes mellitus is a disease with a constant, significant increase in the number of patients, low quality of life and high medical care costs. In this moment we have many diabetes patients with poor control of blood glucose. Some authors consider that the closed-loop system, (artificial pancreas) is the best solution. An extra-corporeal blood glucose sensor is coupled to a computer, which controls the rate of infusion of insulin so as to maintain normal glycaemia. This paper presents the structure of predictive algorithm that can remove most erroneous values. In addition, several numerical simulation results are given where the proposed predictive method outperforms well-known average and median voters. The results of this study can be also applied, to other physiological systems; it offers important data for the medical practice. These findings may have significant clinical implication in diagnosis of the diabetes mellitus, in blood glucose monitoring and in the management of the diabetes therapy.
Keywords: Continuous glucose monitoring, exponential smoothing method, insulin pump, control of blood glucose.

Anca-Loredana ION*
*University of Craiova, Faculty of Automation, Computers and Electronics, Romania (phone: 40-251-438-198; fax: 40-251-438-198; e-mail: aion@software.ucv.ro).

Abstract: This paper is a part of a complex study for automatic detection of medical image diagnostic. Thus, a method based on fuzzy clustering model is developed for medical image segmentation. At the end of this process, an image is represented as a collection of color regions and the visual features are mapped to semantic indicators. After that, the semantic rules are learned by applying the Apriori algorithm on a collection of diagnosed medical images. The experiments were realized on a medical database, which contains images obtained through different medical proceedings: endoscopy, radiology, magnetic resonance imaging etc.
Keywords: Image recognition, image segmentation, image color, image texture, image clustering.

Ioan V. LEMENI*
* University of Craiova, Craiova, Romania (Tel: 0251-435666; e-mail: ioan.lemeni@ cs.ucv.ro)

Abstract: In real classification applications the patterns of different classes often overlap. In this situation the most appropriate classifier is the one whose outputs represent the class conditional probabilities. These probabilities are calculated in traditional statistics indirectly, in two steps: first the underlying prior probabilities are estimated and then the Bayes rule is applied. Popular methods for density estimation are Parzen Window and Gaussian Mixture. It is also possible to calculate directly the class conditional probabilities using the logistic regression, k-Nearest Neighbours algorithm or a Multilayer Perceptron Artificial Neural Network. Many methods, direct or indirect, perform poorly when the underlying prior probability densities are discontinuous along the support's border. This paper will present a method for detecting the discontinuity by analyzing samples drawn from the underlying density. Knowing the densities are discontinuous will help to choose an estimator insensitive to discontinuities.
Keywords: probability density function estimation, class conditional probability estimation.

Marius MARIAN*, Andrei PÎRVAN**
*Department of Computer Science and Information Technology, University of Craiova, RO 200440, Craiova, Romania (Tel: +40-251-438.198; e-mail: marius.marian@cs.ucv.ro)
**University of Craiova, Romania, RO 200440 (e-mail:andrei.pirvan@yahoo.com)

Abstract: These paper details the actual problems encountered while using digital certificates with actual web browsers and mail user agents. After setting up a public-key infrastructure for the community of users within the University of Craiova, we have begun testing the issued digital certificates on different usage scenarios. The personal security profile of a subscriber consists typically of its private key, the corresponding digital certificate, and the chain of hierarchically-related certification authorities' certificates. We have focused on the preliminary step of installing public-key certificates within the secure-credentials repositories of today's most commonly used browsers. The test cases were devised for some of the most popular operating systems employed by the users of our community.
Keywords: public-key infrastructures, digital certificates, security managers, web browsers.

Ileana-Diana V.D. NICOLAE*, Petre-Marian T. NICOLAE**, Dorina-Mioara C. PURCARU***
*Department of Computers and Information Technology (e-mail:nicolae_ileana@software.ucv.ro)
**Department of Electrical Engineering, Power Systems and Aeronautics (e-mail:pnicolae@elth.ucv.ro)
***Department of Automation, Electronics and Mechatronics (e-mail dpurcaru@electronics.ucv.ro), University of Craiova, Decebal Blvd. 107

Abstract: The paper is dedicated to computational aspects related to the Discrete Wavelet Transform use for the analysis of electric signals in power systems. The most important aspects considered when selecting the analysis method are related to the use of proper techniques, able to avoid (or at least to limit as much as possible) the obtaining of inaccurate results because the applied theoretical algorithms assume signals of infinite length whilst only finite segments of them are submitted to the computational process at a specific moment. One intends to obtain reliable power quality indices (with minimum deviations from those calculated with the Fast Fourier Transform) and non-(or minimum) delayed, non-ambiguous "response" of details vectors to disturbances. The avoiding of fake faults detection or of real ones' missing, owing to the "edge effect", along with other important aspects concerning the real-time operation (execution time and memory consumptions) were also considered.
Keywords: Signals processing, wavelet analysis, quality in power systems.

Ionut-Aurelian NISIPEANU*, Elena BUNCIU (STANCIU)**, Roxana STANICA***
*Department of Computer Science and Information Technology, University of Craiova, RO 200440, Craiova, Romania (Tel: +40-251-438.198; e-mail: marius.marian@cs.ucv.ro)
**University of Craiova, Romania, RO 200440 (e-mail:andrei.pirvan@yahoo.com)

Abstract: In this paper some bioprocesses parameters control methods in the case of a continuous BIOSTAT A PLUS bioreactor are presented and analysed. More exactly, three experiments regarding the control of temperature, pH and pO2 are developed and analysed. For the temperature control it was used a PID controller whose parameters were tuned by using Nichols method and for the stirrer speed control to adjust the controller parameters the Ziegler-Nichols method was used. In the case of pH control the parameters of a PI controller were designed and tuned by using the Hokushin method. All the three tuning control methods were confirmed by practical experiments.
Keywords: Bioprocesses, bioreactors, temperature control, pH control, pO2 control.

Andreea Raluca OLTEANU*, Ionel-Cristinel SIMION**, Alina Gabriela PURCAREANU***, Nicu-George BÎZDOACÂ****
* Automatics Department, Faculty of Automation, Computers and Electronics, Craiova, Romania, (e-mail: a_raluk@yahoo.com)
** Mechatronics Department, Faculty of Automation, Computers and Electronics, Craiova, Romania, (e-mail: cristisimion84@gmail.com)
*** Mechatronics Department, Faculty of Automation, Computers and Electronics, Craiova, Romania, (e-mail: alinapurcareanu@yahoo.com)
**** Mechatronics Department, Faculty of Automation, Computers and Electronics, Craiova, Romania, (e-mail: nicu@robotics.ucv.ro)

Abstract: Research in the area of computational approaches to emotions has received increased attention over the last years. The present paper focus two directions: first part investigate emotion from a computational perspective, in order to improve human - robot interaction. The second part presents an experimental approach regarding animatronics design in conjunction with realistic animated motions. The protocol implemented source files are indicated as download files and future project development directions are presented.
Keywords: Emotion, social robot, emotional behavior, computational model of emotion, human - robot interaction.

Teodoru R. POPA*, Andrei C. MOCANU*
*University of Craiova, Faculty of Automation, Computers and Electronics, RO-200440 Romania (e-mail: tpopa@software.ucv.ro).

Abstract: Since the implementation of the emergency coronary care units in the 1960s, ECG monitoring devices were used to diagnose and monitor major life-threatening arrhythmias in patients with cardiac disease. Currently, ECG monitoring plays a major role in other hospital departments as intensive therapy units, operating rooms and emergency rooms. ECG monitoring use has expanded from simple cardiac rhythm detection to complex heart analysis like the detection of complex arrhythmias, atrial fibrillation T wave abnormalities, ST-segment elevation in cardiac ischemia and its role is still expanding. We consider there is a critical medical requirement for continuous development of modern medical devices like ECG that allows doctors not only access to better medical systems, but also to customize access and usage of the medical data. In this article we explain the implementation of an open-source ECG system to visualize XML files obtained from GE MAC5500 ECG machines, without using their specific platform, and a way to automatically detected heart abnormalities based on ECG. It deals with reading, displaying and processing data from a medical electrocardiograph.
Keywords: Data visualization, XML, electronic medical record (EMR), electrocardiogram (ECG), vectorcardiogram (VCG).

Cosmin M. POTERAS*, Calin CONSTANTINOV*, Mihai L. MOCANU*
*University of Craiova, Faculty of Automation, Computers and Electronics, RO-200440 Romania (e-mail: {cpoteras, mmocanu}@software.ucv.ro)

Abstract: Computational steering aims both to interfere with an otherwise autonomous computational process, to change its outcome, and to enable the discovery of new features of the computational processes through integrated experiments. Traditionally, computational steering has been applied to large, compute-intensive and non-interactive simulations, where it more specifically refers to the practice of guiding a simulation experiment into some region of interest. In this paper, we review the motivations for computational steering and introduce an evolutionary design for a framework that takes into consideration two of its main important aspects, program steering and data steering, together with an approach for static scheduling based on a genetic algorithm. We then outline the capabilities of the framework by simulating the execution for three categories of applications, with low, medium and high communication needs, under two running scenarios - with and without tasks migration (required remote data will always be transferred). The results showed that program steering could bring more benefits in the given setting than data steering, and that there is a reasonable loss of efficiency between 16 and 64 processors, which could be explained correlated to the loss in data transfers gain.
Keywords: Computational steering, modeling, simulation, visualization, genetic algorithm.

Viorel-Adrian STANCIU*, Nicu George BÎZDOACA*
* University of Craiova, Automatic Control Department, A.I. Cuza Str. No. 13, RO-200585 Craiova, Romania, E-mail: {stanciu, nicu}@robotics.ucv.ro

Abstract: The present study investigates actuator properties of ElectroActive Polymers - EAP - like artificial muscle. After presenting a brief introduction on the structure and experimental applications of the artificial muscle, the paper aims at highlighting the experimental results on the response of Ionomeric Polymer-Metal Composite (IPMC) EAP to the application of a bipolar and rectangular signal type. In order to develop a realistic solution for bionic structure, the experiments are oriented to finger like structure. The present paper explore the dynamic model of two link finger structure as tendon type actuation. The results offer a strong background for future research in the use of EAP IPMC in robotics and mechatronics.
Keywords: Robotics, mechatronics, electroactive polymer (EAP), ionomeric polymer-metal composite (IPMC).

Mihai STANCU*
*Faculty of Mathematics and Computer Science, Department of Computer Science, University of Craiova, Romania (e-mail: mihai.stancu@yahoo.com)

Abstract: XML is getting very popular as data exchange standard format in web applications due to its portability and information exchange features. When applications must communicate with each other or must integrate information from several data sources, this standard is very useful. However, querying the XML data raise some efficiency problems because of its hierarchical structure and text representation. Over the last years many researches have been done to establish the optimal way of storing XML data in a relational model for taking advantage of the query processing abilities that a relational model can bring. Thus, XML databases have emerged trying to respond to those needs. When we speak of a higher level application, ECM Systems are the standard for developing applications that manage all the unstructured electronic information inside an organization. Over the past years a new information exchange standard has been developed, called CMIS, that unify the data exchange format between the existing ECM Systems. This paper shows how the CMIS standard can be extended to make possible the information exchange operations between ECM Systems and XML Databases. There is also proposed a CMIS interface implementation for a particular XML database, namely EMC xDB. Some performance tests of this implementation have been done. Keywords: XML, information exchange, document management.

Florin STÎNGA*
* University of Craiova, Department of Automatic Control, 13 A.I. Cuza str., Craiova, Romania ( e-mail: florin@automation.ucv.ro)

Abstract: In this paper the problem of stabilization of a particular class of hybrid systems is approached. For the class of discrete-time hybrid systems with time-delays, it is shown that under some assumption it is possible to express these systems in a common representation: piecewise affine systems. The stability of such systems is proved through Lyapunov theory, more accurate through quadratic, multiple Lyapunov functions. For stabilization task we design a piecewise linear state feedback control law. Finally, some simulation results are presented.
Keywords: Discrete-time hybrid systems with time-delays, piecewise affine systems, multiple Lyapunov functions.

Viorel STOIAN*
* University of Craiova, Faculty of Computers, Automation and Electronics, Automation, Electronics and Mechatronics Department, Craiova, Roumania, (Tel: +40 (0)746-11.70.78; e-mail: stoian@ robotics.ucv.ro)

Abstract: This paper presents a faster method for robot modelling by direct kinematics using "Complete Kinematical Structure - CKS". It is a structure with 3 revolute joints and 3 prismatic joints which can assure a complete positioning and orientation in space for an object attached to its terminal. Its homogeneous transformation operator between fix frame (RF) and mobile frame (RM) is determined. An algorithm for robot modelling by CKS is enunciated. This structure is a useful tool for a faster establishing of the kinematical model (direct problem) for the robot arms and helps to resolve the inverse kinematical problems in some situations. By classical methods we obtain the same operator if we have the same position and orientation for the frames R0 (fix) and RM (mobile). Using this procedure, a great part of tedious and time-consumer operations of matrix multiplication (present in classical method) is eliminated. Finally, two applications of some robotic structures are presented.
Keywords: Kinematical model, complete kinematical structure, modelling algorithm, DH method.

Urziceanu Ionut*, Pablo Varona**, Francisco de Borja Rodriguez**, Juan Gonzalez-Gomez***, Fernando Hererro Carron**, Mircea Nitulescu****
* Department of Automation, Electronics and Mechatronics, University of Craiova, Craiova, Romania (e-mail: urziceanuionut@yahoo.com)
** Autonoma University, Madrid, Spain(e-mail: pvarona@uam.es)
*** Universidad Carlos III, Madrid, Spain (e-mail: juan@iearobotics.com)
**** Department of Automation, Electronics and Mechatronics, University of Craiova, Craiova, Romania (e-mail: mirceanitulescu@robotics.ucv.ro)


Abstract: Bio-inspired neural circuits that fall in the category of central pattern generators have been used mainly in research concerning autonomous locomotion for legged and serpentine robots. This paper presents a novel locomotion controller based on bursting neuron models and time evolving synapses that is used to control a differential wheeled robot. We show how the proposed central pattern generator model is able to generate a flexible yet robust rhythm which is used to sequentially drive the wheels in a situation where the joints can only partially rotate.
Keywords: Central pattern generator, bio-inspired, wheeled robot, autonomous locomotion controller.