Catálogo de publicaciones - libros

In this paper we propose a generic approach to acquire navigation skills for nonholonomic vehicles in unknown environments. The algorithm uses reinforcement learning to update both the vehicle model and the optimal behaviour at the same time. After the training phase, the vehicle is able to explore the environment through a wall-following behaviour. The vehicle can also reach any goal position by the virtual wall concept. The method does not require function interpolation to obtain a good approximation to the optimal behaviour. The learning time was only a few minutes to acquire the wall-following behaviour. Both simulation and experimental results are reported to show the satisfactory performance of the method.

Morphologically independent vectors correspond to approximations to the vertices of the convex hull covering the data vectors in high dimensional space. We use Morphological Associative Memories (MAM) for the induction of sets of morphologically independent vectors from data. Simultaneous Localization and Mapping (SLAM) is the process of simultaneously building a map of the environment and localizing the mapping agent. In this paper we explore the realization of non-metric SLAM using a visual information based approach relying on morphologically independent images induced from a mobile robot camera image stream. The selected images are proposed as the landmarks for localization, building simultaneously a qualitative map of the environment. We report results of some experiments on data gathered from an indoor ambient.

In this work, Self Organizing Map (SOM) is used in order to detect and classify the broken rotor bars and misalignment type mechanical faults that often occur in induction motors which are widely used in industry. The feature vector samples are extracted from the sampled line current of motors with fault and healthy one. These samples are the poles of the AR model which is obtained from the spectrum of sampled line current. The waveforms are obtained from four different 3 hp test motors. Two of them have different number of broken rotor bars, one test motor has misalignment problem and the last one is the healthy motor. Broken rotor bar and misalignment faults are successfully classified and distinguished from the healthy motor using SOM classification with the feature vectors. It is also worth to mention that discrimination of different number of broken rotor bars has been achieved.

The topology verification is an important problem in power system engineering. The paper presents the solution for this problem with use of the method, that is independent of state estimation. The method combines utilization of knowledge about power system and radial basis function networks. It allows to perform the power system topology verification as a series of verification processes for particular nodes of a power network. In the paper the possibility of such decomposition is proved. A principle of the method is described. Next, the computational example of topology verification with use of the characterized method is presented. At the end, the features of the method are analyzed, paying special attention to its efficiency.

Real-life applications of intelligent systems that use neural networks require a high degree of success, usability and reliability. Power systems applications can benefit from such intelligent systems; particularly for voltage stabilization. Voltage instability in power distribution systems could lead to voltage collapse and thus power blackouts. This paper presents an intelligent system which detects voltage instability and classifies voltage output of an assumed power distribution system (PDS) as: stable, unstable or overload. The novelty of our work is the use of voltage output images as the input patterns to the neural network for training and generalizing purposes, thus providing a faster instability detection system that simulates a trained operator controlling and monitoring the 3-phase voltage output of the assumed PDS. Experimental results suggest that our method performs well and provides a fast and efficient system for voltage instability detection.

This work introduces an artificial neural network based speed control system for AC induction motors. In motor control, Radial Basis Function (RBF) network and Volt-Hertz (V/f) method have been adopted as neural controller and scalar driving tool respectively. V/Hz method has been preferred for the sake of simplicity and well-known reliability. The proposed control scheme consists of two main parts; a RBF network and a reference model unit. RBF based main controller supplies appropriate control signals to the driving unit to compensate nonlinearity of the induction motor and track the reference model output. Reference model, is very stable linear filter which is supplying set values to be imitated by induction motor. The success of the proposed control scheme has been demonstrated by experimental results; induction motor has been able to track the prescribed speed trajectory with rather small errors and good stability under properly loading conditions.

In this paper we explore the interest of computational intelligence tools in the management of heterogeneous communication networks, specifically to predict congestion, failures and other anomalies in the network that may eventually lead to degradation of the quality of offered services. We show two different applications based on neural and neurofuzzy systems for Quality of Service (QoS) management in next generation networks for V2oIP services. The two examples explained in this paper attempt to predict the communication network resources for new incoming calls, and visualizing by means of self-organizing maps the QoS of a communication network.

In this paper, intrusion detection method using Bayesian Networks was estimated probability values of behavior contexts based on Bayes theory and Indirect relation. The contexts of network-based FTP service was represented Bayesian Networks of graphic types. We profiled concisely network-based FTP behaviors using behavior context by prior, posterior and Indirect relation. And this method be able to visualize behavior profile to detect/analyze anomaly behavior by BF-XML. We achieve simulation to translate audit data of network into BF-XML which is behavior profile of semi-structured data type for anomaly detection and to visualize BF-XML as SVG.

Spam, also known as Unsolicited Commercial Email (UCE) is becoming a nightmare for Internet users and providers. Machine learning techniques such as the Support Vector Machines (SVM) have achieved a high accuracy filtering the spam messages. However, a certain amount of legitimate emails are often classified as spam (false positive errors) although this kind of errors are prohibitively expensive.

In this paper we address the problem of reducing particularly the false positive errors in anti-spam email filters based on the SVM. To this aim, an ensemble of SVMs that combines multiple dissimilarities is proposed. The experimental results suggest that the new method outperforms classifiers based solely on a single dissimilarity and a widely used combination strategy such as bagging.

In this paper the utility of using the Self Organizing Maps (SOM), in conjunction with U-matrix, to visualize the evolution of a social network community formed by a set of blogs is shown. Weblogs are dynamic websites updated via easy-to-use content management systems whose links tend to mirror or in some cases establish new types of social relations, thereby creating a social network. Analyzing the evolution of this network allows the discovery of emerging social structures and their trends in growth. Here we apply this method to Blogalia, a blog hosting site from which we have a complete set of data. The proposed procedure not only gives some insight on how communities form and evolve, but would also enable to predict the future paths that their members will take.