Search

Researches related Autonomous Underwater Vehicles (AUVs) have boomed in recent years mainly due to the development of sensors and computers technologies. In Cuba the Group of Automation, Robotics and Perception and the Hydrographic Research Center have been working on the development of an AUV based on low-cost hardware architecture. In this paper the design of a navigation system based on a passive observer is shown. The algorithm consists of a nonlinear model of the vehicle in combination with linear approximations of ocean waves and currents. The tuning procedure is relatively easy and the observer obtained is feasible to implement in real time. The simulations performed using MatLab/Simulink show good results in terms of disturbances rejection and estimation of navigation parameters needed like velocities and positions. [ABSTRACT FROM AUTHOR]

The purpose of the optimum designs is to determine the optimal experimental conditions in terms of minimum variance of the estimated vector of parameters, so that statistical inferences can be generated as accurate as possible. This theory presupposes knowledge of the function relating the explanatory variables with the response variable. In this situation, usually get designs with support points as parameters in the model. Since p-point designs assume that the model function is known, these could not be optimal in some practical situations because they do not allow to test the goodness of fit of the assumed model. In this paper we present a generalization of the methodology proposed by to increase the number of support points in D-optimality criterion. An expression for the variance of the predicted response in terms of a constant δ will be provided, which allow to determine the points to be added to the D-optimal design. We propose a strategy for choosing δ. Finally we provide a pseudo-optimal design with (p + s) support points. [ABSTRACT FROM AUTHOR]

In this paper we propose a new methodology for the prediction of nonlinear time series using genetic programming. The proposed approach is based on incorporating the concept of functional blocks and the modification of the genetic algorithm so that it operates with it. The functional blocks represent well known statistical models for the time series forecasting. The proposed algorithm allows the exploration and exploitation of regions where there is greater possibility of finding better forecasting models. Two Benchmark time series were predicted in order to validate the proposed approach, and it was found that our methodology predicts more accurately the time series considered, in comparison with other nonlinear models. [ABSTRACT FROM AUTHOR]

Growth analysis with non-linear regression models (NLM) allows deriving parameters (GP) and indicators (GI) of growth under specific production conditions that can be taken into account in genetic improvement programs. The objectives of this study were to select and adjust a NLM to the growth of Tropicarne cattle raised in the Gulf of México coastline. The NLM evaluated were Brody, Logistic, Bertalanffy and Gompertz. The growth information analyzed corresponded to 12 890 monthly weigh-ins, from birth to 24 months of age, in 1787 animals. The GP estimated were adult weight (AW) and maturity rate (MR); and the GI, age (API) and weight (WPI) at point of inflection, age at 50 % of maturity (A50M) and percentage of maturity at 18 months of age (PM18). The analyses were carried out with the NLIN procedure of SAS. The selection of models was done based on the average prediction error, variance of the prediction error, multiple and adjusted determination coefficient, the Durbin-Watson statistic, and the Akaike information criterion. The Brody model, followed by the Bertalanffy model, showed a better adjustment with estimations of 467.5 and 378.8 kg for AW, and 0.0612 and 0.132 kg kg-1 month-1 for MR. Derivations of the API, WPI, A50M and PM18 were 3.8 months, 112.2 kg, 7.4 months and 85 %. The Brody model allowed a better description of the growth of Tropicarne cattle, which implies slow MR, characteristic of the growth of bovines in the tropics, in contrast with the typical sigmoid curve for some breeds in temperate climates. [ABSTRACT FROM AUTHOR]

Theoretical extension of the linear moving average models to the nonlinear case is direct and straightforward. However, the practical use of nonlinear moving average models is limited because of complexity of the parameters space and the impossibility of establishing the analytic derivate of estimation function. In this article, we evaluate the use of the Differential Evolution – Particle Swarm Optimization hybrid algorithm for calculating the optimal parameters of the nonlinear moving averages model. Obtained results show that the used meta-heuristic technique is able to calculate with more accuracy the values of the parameters of the model in comparison with the traditional algorithms. This finding encourages us to explore the use of meta-heuristics in the estimation of the parameters of other nonlinear moving average models. [ABSTRACT FROM AUTHOR]

This paper is about developing a nonlinear model to predict the behavior of future exchange rate based on the opinion of the economic agents participating in the dollar/peso market. Such views are treated with Fuzzy Logic and a variant of it, known as the Theory of Forgotten Effects. The aim is to find a mechanism for making coverage decisions that allow us an optimal exchange rate risk management at a lower cost than that which involves operations with traditional hedging instruments. For the period of investigation and applying this model, the results support that the collective opinions of economic experts involved in the decision making risk management of exchange rate provide better results than those using traditional methods in the future markets. [ABSTRACT FROM AUTHOR]

Agent-based modeling is a new research method that allows the complexity, emergency and nonlinearity of many social processes to be managed in a simple way. In this paper we define agent-based modeling — and its different classes— on the basis of a classic example: the Schelling model of urban segregation, which seeks to explain how it is possible that ghettos can arise in cities where citizens have no racial prejudices. Finally, some of its possible applications are considered: developing, formalising and assessing sociological theories; deciding amongst rival theories about the same social phenomenon; integrating different theories in a common setting; explaining social dynamics through their micro-mechanisms; carrying out experiments in a virtual world; assessing social policies and their a priori results; and predicting certain social phenomena. [ABSTRACT FROM AUTHOR]

In this communication, a critical bibliographical survey on discrete least squares (LS) presentation is firstly made, and especially of data fitting (DF). Afterwards, to help to overcome part of the limitations and deficiencies that were shown in the bibliography, a formal presentation of generalized least squares (GLS, in general non linear by default) is made, and very especially of DF for vector valued functions (VVF); the corresponding non linear system of equations, extension of the well known normal equations, is derived. From there, some particular cases are obtained, for example, when the function is real valued, and when the approximation is linear. This second example is just linear generalized least squares (LGLS). [ABSTRACT FROM AUTHOR]

The article analyses the main trends and concepts from complexity studies. It acknowledges unresolved problems in Public Management and it presents an approach to their modeling, lineal, non lineal, static and dynamic. It is concluded that the administration of organizations faces complex phenomena, and requires models that can reflect this complexity. The extreme simplification of these phenomena is risky, and it requires a profound analysis of the mathematical modeling of Social Sciences and Administration. It is necessary to work with multi-disciplinary teams and to recover systematic modeling, as well as, problem solving based on heuristic logic. [ABSTRACT FROM AUTHOR]

We indicate how to handle a large database consisting of nonlinear time series, applying different nonlinear modelling techniques to this kind of times series. Nowadays in the current references there is no explicit guide of how to manipulate data from nonlinear time series; however, there are approaches that can be taken account. To this end we studied a monthly database corresponding to South Oscillation Index (SOI) and between the years 1866 to 2006. It explains how there must manipulated this database whose data possess nonlinear characteristic, which will be used to do forecasts several steps ahead. Two standard tests to this database were applied, the Average Mutual Information (AMI) and the False Nearest Neighbours (FNN). The optimal spacing of the information was obtained as well as the number of values backward necessary to predict values towards the future. Then, several models were designed of artificial neural nets (ANN), with different learning rules, function of transfer, elements of process (or neurons) in the hidden layer, etc., that allowed to do forecasting of up to 20 steps ahead. The best networks were those that possessed the rules of learning called extDBD and Delta-Rule, and sigmoid as well as hyperbolic tangent as function of transfer. The type of used network was one of feedforward multilayer perceptron and trained by means of backpropagation technique. Networks were proved by one, two hidden layers and without any hidden layer. The best model that was obtained it turned out to be one that consisted with an alone hidden layer. [ABSTRACT FROM AUTHOR]

A computational strategy is presented for the least squares curve fit of nonlinear models which allows obtaining a good initial approximation that guarantees convergence of the iterative process required by its solution. This strategy is illustrated with biological models. [ABSTRACT FROM AUTHOR]

Time series prediction is an important research problem due to its implications in engineering, economics, finance and social sciences. An important topic about this problematic is the development of new models and its comparison with previous approaches in terms of forecast accuracy. Recently, support vector machines (SVM) have been used for time series prediction, but the reported experiences are limited and there are some problems related to its specification. The aim of this paper is to propose a novel technique for estimating some constants of the SVM usually fixed empirically by the modeler. The proposed technique is used to estimate several SVM with the aim of forecast five benchmark time series; the obtained results are compared with the statistics reported in other papers. The proposed method allow us to obtain competitive SVM for the time series forecasted in comparison with the results obtained using other most traditional models. [ABSTRACT FROM AUTHOR]

This work describes and analyses the Extended Kalman Filter with regard to an equation that relates the dynamic of the expected real interest rates and inflation. An estimation of the ex ante inflation for a preestablished data set is carried out. This is compared with the same calculation using a moving horizon estimation method for situations that are non lineal due to parametric uncertainties of the model. From the application of these methods to real data, it can be concluded that the estimations based on the moving horizon method, combined with a heuristic optimization algorithm, yield better results. [ABSTRACT FROM AUTHOR]

Global support radial base functions are effective in eliminating impulsive noise in non-linear processing. This paper introduces an interpolation technique which efficiently reduces image impulsive noise by means of an interpolant obtained through radial base functions. These functions have been used in a research project designed to develop a system for the recovery of images of Amazonian aquatic resources. This technique starts with the tagging by interpolation of noisy image pixels. Thus, a value of reconstruction for the noisy pixels is generated using neighboring pixels. The results obtained with this technique have proved comparable and often better than those obtained with previously known techniques. According to results analysis, this technique can be successfully applied on images with high noise levels. The results are low error in noisy pixel reconstruction and better visual quality. [ABSTRACT FROM AUTHOR]

A novel algorithm is presented for the optimal estimation of parameters and for determining the initial conditions in non linear mathematical models. The work was developed based on models derived from property balance equations. A matrix relation between the variation of degrees of freedom of the process and the variations of the input variables was used to formulate the objective function, which was optimized via orthogonal projection methods. The technique was applied to a continuous stirred tank reactor obtaining results that were consistent with the dynamic behavior of the reactor. [ABSTRACT FROM AUTHOR]

This article presents the results of empirical assessments of the nominal exchange rate in Colombia, conducted using linear specifications (distributed lag model) and non-linear specifications (artificial neural networks and neural diffusion systems). All of the econometric specifications employ order flow as the determinant for the exchange rate, the most representative variable for models that attempt to explain the nominal exchange rate, supported by the exchange market microstructure. The forecast error measurements used to assess the predictive ability of the models evidenced that linear and non-linear models had an advantage over a random path models, attributable to the order flow. Non-linear models were somewhat superior to the linear model in terms of the root mean square error, Theil's coefficient, and the percentage of predicted correct changes, but the difference was so slight that it does not enable confirming the hypothesis that the exchange rate generation process is non-linear. [ABSTRACT FROM AUTHOR]

This work examines the relationship between predictability of return sign and conditional moments. A nonlinear time series model, later restricted to a first-order Taylor expansion in the innovations, is used for this purpose. As a result, it is shown why sign predictability depends only on odd-numbered order conditional moments (on the conditional mean in the restricted model), which makes testing for linearity or non linearity in the mean interesting before attempting directional forecasting. As an application, the presence of nonlinearity in the mean in the Colombian stock exchange index IGBC is examined applying the BDS test to the residuals of an ARMA-GARCH filter and the White Neural Network Test for Nonlinearity to the residuals of an AR filter. Both tests show that the IGBC exhibits nonlinearity in the mean. [ABSTRACT FROM AUTHOR]

This paper analyzes the use of the oscillation-based test (OBT) to test high-order continuous-time ladder filters. The OBT implementation requires converting the filter under test into a non-linear oscillator. The prediction of the oscillation parameters (frequency and amplitude) is done using a combination of analytical and graphical methods, resulting very appropriate for the filters under study (fifth-order elliptic and Butterworth, RC active type). For characterizing the OBT scheme, a single deviation fault model for the filter passive components was adopted, and the lowest deviations in the value of the components, capable of being detected by monitoring amplitude and frequency, were established. The fault simulation results show that OBT is very appropriate for detecting faults in the Butterworth filter. For the elliptic filter, good results were observed if the hard-to-test faults are related to the filter behavior. [ABSTRACT FROM AUTHOR]

A dynamic simulator for vacuum batch crystallization processes using a non-linear model is presented. The model is based on the equation of balance population, on mass and energy balances inside the crystallizer, on the steam continuity equation for the heating jacket and on constitutive relations. Using the technique of crystal size distribution moments, the equation of balance population is converted into a set of differential equations. The construction of simulator was done using Visual Basic. An example of its use is proposed comparing results with those of another simulator and with experimental information obtained in a pilot equipment pilot, obtaining similar results. [ABSTRACT FROM AUTHOR]

Recent progress obtained in semiclassical approximations to the propagator of the Wigner function [Phys. Rev. Lett. 96, 070403 (2006)] indicate that the phase space representations of quantum dynamics is a promising strategy to surmount the long-standing problems of semiclassical propagation. In this work we apply the semiclassical Wigner propagator to the Morse oscillator, a benchmark system for quantum-mechanical propagation in strongly nonlinear potentials. In particular we compare on the van Vleck propagator to the exact quantum solution, applied to a coherent state as initial condition. [ABSTRACT FROM AUTHOR]

Presently paper deals with stability analysis in structural elements. The considerations of lineal and non lineal analysis of stability are exposed and it is exemplified with a finite element structure model, settling down interesting conclusions. [ABSTRACT FROM AUTHOR]

This paper analyzes the evidence of non-linearities in the reaction function of the Central Bank of Chile as it relates to its final objective, inflation, and/or to its intermediate objectives defined by the monetary policy transmission channels. The study uses the flexible approach to non-linear inference proposed by Hamilton (2001), which allows the estimation of non-linear functions by using the data to determine the specific form of the non-linearity to be adopted instead of making any prior assumption of a particular functional form. It is observed that even though the Central Bank's response is linear in relation to the deviations of the inflation rate from its target, there is evidence of non-linearities with respect to the business cycle. Contrary to previous studies, in particular Corbo (2002), the current account gap does not have any predictive role in the reaction function for the period being analyzed. [ABSTRACT FROM AUTHOR]

We describe here electrothermal poling, which is one of the techniques most commonly used to induce nonlinear responses in materials such as fused silica, along with the experimental procedure we have used for (SHG) once nonlinearity has been induced. We present some evidence regarding low intensity wide band transmittance measurement inhability to detect refraction index changes due to nonlinearity induction. [ABSTRACT FROM AUTHOR]