Entorn de programació pel robot Stäubli mitjantçant MatLab, aplicació a un sistema de recol·lecció de peces basat en visió artificial

Els objectius del projecte són: realitzar un intèrpret de comandes en VAL3 que rebi les ordres a través d’una connexió TCP/IP; realitzar una toolbox de Matlab per enviar diferents ordres mitjançant una connexió TCP/IP; adquirir i processar mitjançant Matlab imatges de la càmera en temps real i detectar la posició d’objectes artificials mitjançant la segmentació per color i dissenyar i realitzar una aplicació amb Matlab que reculli peces detectades amb la càmera. L’abast del projecte inclou: l’estudi del llenguatge de programació VAL3 i disseny de l’ intèrpret de comandes, l’estudi de les llibreries de Matlab per comunicació mitjançant TCP/IP, per l’adquisició d’imatges, pel processament d’imatges i per la programació en C; el disseny de la aplicació recol·lectora de peces i la implementació de: un intèrpret de comandes en VAL3, la toolbox pel control del robot STAUBLI en Matlab i la aplicació recol·lectora de peces mitjançant el processament d’imatges en temps real també en Matlab

Modulation of inflammatory pathways by the HIV protease inhibitors

Les inhibiteurs de la protéase du VIH (IP) constituent une des classes de traitements antirétroviraux parmi les plus utilisés au cours de l'infection par le VIH. Leur utilisation est associée à divers effets secondaires, notamment la dyslipidémie, la résistance à l'insuline, la lipodystrophie et certaines complications cardio-vasculaires. Ces molécules ont également des propriétés anti-tumorales, décrites chez des patients non infectés par le VIH. Pourtant, les mécanismes moléculaires à l'origine de ces effets annexes restent méconnus. Dans ce travail, nous démontrons que les IP, comme le Nelfinavir, le Ritonavir, le Lopinavir, le Saquinavir et l'Atazanavir, entrainent la production d'interleukine-lß (IL-lß), une puissante cytokine pro-inflammatoire, connue pour son rôle central dans les maladies inflammatoires. La sécrétion d'IL-lß requiert la formation de l'inflammasome, un complexe protéique intracellulaire servant de plateforme d'activation de la caspase-1 et, par la suite, à la maturation protéolytique de certaines cytokines, dont l'IL-lß. Dans les macrophages murins en culture primaire, ainsi que dans une lignée de monocytes humains, nous démontrons que les IP augmentent la maturation et la sécrétion de l'IL-lß via l'induction d'un inflammasome dépendant de ASC. De plus, nous établissons que les IP induisent spécifiquement l'activation de AIM2, un inflammasome détectant la présence intracytosolique d'ADN viral ou bactérien. Nos résultats démontrent l'existence d'une nouvelle voie d'activation de l'inflammasome AIM2 par un signal endogène dont la nature reste à définir. Ces données suggèrent que AIM2 pourrait jouer un rôle important dans la promotion de l'activité anti-tumorale ainsi que dans les autres effets annexes observés chez les patients traités par IP. -- HIV protease inhibitors (Pis) are among the most often used classes of antiretroviral drugs for HIV infection. Treatment of patients with HIV-PIs is associated with the development of metabolic side effects including dyslipidemia, insulin resistance, lipodystrophy and cardiovascular complications. In addition, these drugs have been reported to have anti¬tumoral properties in non-infected patients, however the molecular mechanisms causing these off-target effects are still unclear. Here we show that the HIV-PIs, such as Nelfinavir, Ritonavir, Lopinavir, Saquinavir and Atazanavir, activate the production of interleukin-lß (IL-lß), a potent pro-inflammatory cytokine that plays a central role in the pathogenesis of inflammatory diseases. The release of IL-lß depends on the activation of the inflammasome, a multiprotein complex that serves as a platform for caspase-1 activation and subsequent proteolytic maturation of cytokines including IL-lß. We found that in mouse primary macrophages as well as in a human monocytic cell line, the HIV-PIs augment the maturation and secretion of IL-lß by triggering an ASC-dependent inflammasome activation. Moreover, we show that the HIV-PIs specifically engage AIM2, a recently characterized inflammasome -forming protein that was described to detect the cytosolic release of bacterial and viral DNA. Our findings demonstrate a new pathway of activation of the AIM2 inflammasome by a yet to be defined endogenous signal and may suggest a possible role for AIM2 in promoting anti¬tumoral activity and off-target effects observed in HIV-PIs treated patients.

Strong minimax lower bounds for learning

Minimax lower bounds for concept learning state, for example, thatfor each sample size $n$ and learning rule $g_n$, there exists a distributionof the observation $X$ and a concept $C$ to be learnt such that the expectederror of $g_n$ is at least a constant times $V/n$, where $V$ is the VC dimensionof the concept class. However, these bounds do not tell anything about therate of decrease of the error for a {\sl fixed} distribution--concept pair.\\In this paper we investigate minimax lower bounds in such a--stronger--sense.We show that for several natural $k$--parameter concept classes, includingthe class of linear halfspaces, the class of balls, the class of polyhedrawith a certain number of faces, and a class of neural networks, for any{\sl sequence} of learning rules $\{g_n\}$, there exists a fixed distributionof $X$ and a fixed concept $C$ such that the expected error is larger thana constant times $k/n$ for {\sl infinitely many n}. We also obtain suchstrong minimax lower bounds for the tail distribution of the probabilityof error, which extend the corresponding minimax lower bounds.

Intellectual property rights and efficient firm organization

I show that intellectual property rights yield static efficiency gains, irrespective oftheir dynamic role in fostering innovation. I develop a property-rights model of firmorganization with two dimensions of non-contractible investment. In equilibrium, thefirst best is attained if and only if ownership of tangible and intangible assets is equallyprotected. If IP rights are weaker, firm structure is distorted and efficiency declines:the entrepreneur must either integrate her suppliers, which prompts a decline in theirinvestment; or else risk their defection, which entails a waste of her human capital. Mymodel predicts greater prevalence of vertical integration where IP rights are weaker,and a switch from integration to outsourcing over the product cycle. Both empiricalpredictions are consistent with evidence on multinational companies. As a normativeimplication, I find that IP rights should be strong but narrowly defined, to protect abusiness without holding up its potential spin-offs.

Major: Electronics and Communication Engineering

Today, information technology is strategically important to the goals and aspirations of the business enterprises, government and high-level education institutions – university. Universities are facing new challenges with the emerging global economy characterized by the importance of providing faster communication services and improving the productivity and effectiveness of individuals. New challenges such as provides an information network that supports the demands and diversification of university issues. A new network architecture, which is a set of design principles for build a network, is one of the pillar bases. It is the cornerstone that enables the university’s faculty, researchers, students, administrators, and staff to discover, learn, reach out, and serve society. This thesis focuses on the network architecture definitions and fundamental components. Three most important characteristics of high-quality architecture are that: it’s open network architecture; it’s service-oriented characteristics and is an IP network based on packets. There are four important components in the architecture, which are: Services and Network Management, Network Control, Core Switching and Edge Access. The theoretical contribution of this study is a reference model Architecture of University Campus Network that can be followed or adapted to build a robust yet flexible network that respond next generation requirements. The results found are relevant to provide an important complete reference guide to the process of building campus network which nowadays play a very important role. Respectively, the research gives university networks a structured modular model that is reliable, robust and can easily grow.

A quantitative analysis of antagonism and inverse agonism at wild-type and constitutively active hamster alpha1B-adrenoceptors.

In order to characterize inverse agonism at alpha1B-adrenoceptors, we have compared the concentration-response relationships of several quinazoline and non-quinazoline alpha1-adrenoceptor antagonists at cloned hamster wild-type (WT) alpha1B-adrenoceptors and a constitutively active mutant (CAM) thereof upon stable expression in Rat-1 fibroblasts. Receptor activation or inhibition thereof was assessed as [3H]inositol phosphate (IP) accumulation. Quinazoline (alfuzosin, doxazosin, prazosin, terazosin) and non-quinazoline alpha1-adrenoceptor antagonists (BE 2254, SB 216,469, tamsulosin) concentration-dependently inhibited phenylephrine-stimulated IP formation at both WT and CAM with Ki values similar to those previously found in radioligand binding studies. At CAM in the absence of phenylephrine, the quinazolines produced concentration-dependent inhibition of basal IP formation; the maximum inhibition was approximately 55%, and the corresponding EC50 values were slightly smaller than the Ki values. In contrast, BE 2254 produced much less inhibition of basal IP formation, SB 216,469 was close to being a neutral antagonist, and tamsulosin even weakly stimulated IP formation. The inhibitory effects of the quinazolines and BE 2254 as well as the stimulatory effect of tamsulosin were equally blocked by SB 216,469 at CAM. At WT in the absence of phenylephrine, tamsulosin did not cause significant stimulation and none of the other compounds caused significant inhibition of basal IP formation. We conclude that alpha1-adrenoceptor antagonsits with a quinazoline structure exhibit greater efficacy as inverse agonists than those without.

Melatonin improves glucose homeostasis and endothelial vascular function in high-fat diet-fed insulin-resistant mice.

Obesity and insulin resistance represent a problem of utmost clinical significance worldwide. Insulin-resistant states are characterized by the inability of insulin to induce proper signal transduction leading to defective glucose uptake in skeletal muscle tissue and impaired insulin-induced vasodilation. In various pathophysiological models, melatonin interacts with crucial molecules of the insulin signaling pathway, but its effects on glucose homeostasis are not known. In a diet-induced mouse model of insulin resistance and normal chow-fed control mice, we sought to assess the effects of an 8-wk oral treatment with melatonin on insulin and glucose tolerance and to understand underlying mechanisms. In high-fat diet-fed mice, but not in normal chow-fed control mice, melatonin significantly improved insulin sensitivity and glucose tolerance, as evidenced by a higher rate of glucose infusion to maintain euglycemia during hyperinsulinemic clamp studies and an attenuated hyperglycemic response to an ip glucose challenge. Regarding underlying mechanisms, we found that melatonin restored insulin-induced vasodilation to skeletal muscle, a major site of glucose utilization. This was due, at least in part, to the improvement of insulin signal transduction in the vasculature, as evidenced by increased insulin-induced phosphorylation of Akt and endoethelial nitric oxide synthase in aortas harvested from melatonin-treated high-fat diet-fed mice. In contrast, melatonin had no effect on the ability of insulin to promote glucose uptake in skeletal muscle tissue in vitro. These data demonstrate for the first time that in a diet-induced rodent model of insulin resistance, melatonin improves glucose homeostasis by restoring the vascular action of insulin.

Assigning proctors to exams with scatter search

In this paper we present an algorithm to assign proctors toexams. This NP-hard problem is related to the generalized assignmentproblem with multiple objectives. The problem consists of assigningteaching assistants to proctor final exams at a university. We formulatethis problem as a multiobjective integer program (IP) with a preferencefunction and a workload-fairness function. We then consider also a weightedobjective that combines both functions. We develop a scatter searchprocedure and compare its outcome with solutions found by solving theIP model with CPLEX 6.5. Our test problems are real instances from aUniversity in Spain.

Adaptive model selection using empirical complexities

Given $n$ independent replicates of a jointly distributed pair $(X,Y)\in {\cal R}^d \times {\cal R}$, we wish to select from a fixed sequence of model classes ${\cal F}_1, {\cal F}_2, \ldots$ a deterministic prediction rule $f: {\cal R}^d \to {\cal R}$ whose risk is small. We investigate the possibility of empirically assessingthe {\em complexity} of each model class, that is, the actual difficulty of the estimation problem within each class. The estimated complexities are in turn used to define an adaptive model selection procedure, which is based on complexity penalized empirical risk.The available data are divided into two parts. The first is used to form an empirical cover of each model class, and the second is used to select a candidate rule from each cover based on empirical risk. The covering radii are determined empirically to optimize a tight upper bound on the estimation error. An estimate is chosen from the list of candidates in order to minimize the sum of class complexity and empirical risk. A distinguishing feature of the approach is that the complexity of each model class is assessed empirically, based on the size of its empirical cover.Finite sample performance bounds are established for the estimates, and these bounds are applied to several non-parametric estimation problems. The estimates are shown to achieve a favorable tradeoff between approximation and estimation error, and to perform as well as if the distribution-dependent complexities of the model classes were known beforehand. In addition, it is shown that the estimate can be consistent,and even possess near optimal rates of convergence, when each model class has an infinite VC or pseudo dimension.For regression estimation with squared loss we modify our estimate to achieve a faster rate of convergence.

Bank versus venture capital

Why do some start-up firms raise funds from banks andothers from venture capitalists? To answer this question,I develop a model of start-up financing when intellectualproperty rights are not well protected. The upside of VCfinancing is that the VC understands the business betterthan a bank. The downside, however, is that the VC maysteal the idea and use it himself. The results of themodel are consistent with empirical regularities onstart-up financing. The model implies that thecharacteristics of the firms financing from venturecapitalists are low-collateral, high-growth and high-profitability. The model also suggests that thetighter protection of intellectual property rightscontributes to the recent dramatic growth of the USventure capital industry.

Model selection and error estimation

We study model selection strategies based on penalized empirical loss minimization. We point out a tight relationship between error estimation and data-based complexity penalization: any good error estimate may be converted into a data-based penalty function and the performance of the estimate is governed by the quality of the error estimate. We consider several penalty functions, involving error estimates on independent test data, empirical {\sc vc} dimension, empirical {\sc vc} entropy, andmargin-based quantities. We also consider the maximal difference between the error on the first half of the training data and the second half, and the expected maximal discrepancy, a closely related capacity estimate that can be calculated by Monte Carlo integration. Maximal discrepancy penalty functions are appealing for pattern classification problems, since their computation is equivalent to empirical risk minimization over the training data with some labels flipped.

Tecnologia VOIP com Asterisk (UM@STERISK)

Este trabalho trata do software livre Asterisk (Elastix), uma central telefónica privada IP ou (PBX IP) que suporta inúmeros protocolos e codecs da tecnologia de voz sobre IP, Abordando a sua instalação, configuração e compatibilidade com hardware de telefonia. O crescimento das redes IP, as técnicas avançadas de digitalização de voz e os mecanismos que permitem a qualidade dos serviços, permitiram a consolidação da telefonia IP. A telefonia IP está em todo-poderoso crescimento, pois além de reduzir os custos das ligações telefónicas, ela permite a ligação entre as redes de dados e de voz, criando uma infra-estrutura única, facilitando a instalação, a manutenção e o gerenciamento. Este trabalho tem como objetivo fazer um estudo da Tecnologia VoIP (Voz sob IP), propor uma estrutura e implementar um ambiente de teste, uma central telefónica VoIP com o uso do Asterisk. A central telefónica VoIP baseada no software livre Asterisk possibilita a interligação de localidades geograficamente distantes uma das outras através das redes IP, sem a necessidade de pagar os altos valores cobrados pelos fabricantes de centrais telefónicas, de hardware proprietário, pela manutenção, fornecimento de equipamentos e licenças, mas permite que se obtenha os mesmos resultados, como por exemplo, que todos os Funcionários e ou colaboradores de uma empresa realizem chamadas telefónicas entre si sem precisar pagar altas taxas cobradas pelas operadoras de telefonia pública. Através da pesquisa bibliográfica sobre a tecnologia VoIP e do estudo do software livre Asterisk, será proposta a implantação da tecnologia através da implementação de um central VoIP. A verificação dessa implantação será realizada através de testes práticos em um ambiente que será desenvolvido.

Aplicação de índice comparativo na avaliação do risco de degradação das terras

A determinação do risco de degradação das terras em uma microbacia hidrográfica constitui importante subsídio para o planejamento agrícola e ambiental. A tendência atual é que as informações necessárias para a avaliação das terras sejam compatíveis com técnicas de geoprocessamento e tratadas de forma menos subjetiva. Todavia, critérios ou parâmetros que avaliem quantitativamente o risco de degradação das terras e que sejam compatíveis com a escala de microbacia hidrográfica e aplicáveis por meio de Sistemas de Informação Geográfica (SIGs) ainda precisam ser mais bem definidos. O objetivo deste trabalho, realizado no segundo semestre de 1996, foi elaborar um método de avaliação comparativa do risco de degradação das terras na bacia hidrográfica dos Marins (Piracicaba, SP), sob diferentes cenários de uso, utilizando SIG. Os mapas de risco de degradação foram gerados com base em matrizes de decisão para os seguintes cenários de uso: (a) sem cobertura vegetal; (b) uso atual da terra; (c) uso único; (d) uso redistribuído; (e) uso planejado. Os resultados foram comparados por um índice denominado "Índice Ponderado de Risco de Degradação (IP D)". Por meio desse índice foi possível comparar quantitativamente o risco de degradação para os cenários de uso propostos. Os resultados mostraram que a tentativa de alterar apenas a distribuição espacial dos usos da terra não foi suficiente para diminuir o risco de degradação final da área, caracterizando, assim, a superutilização das terras da microbacia. Para diminuir os impactos causados por essa superutilização, faz-se necessária a definição de critérios de restrição de ocupação que conciliem áreas de maior suscetibilidade à erosão com usos que ofereçam maior proteção ao solo, levando à diminuição da área ocupada com culturas anuais e semiperenes.

The PPARbeta/delta agonist GW0742 relaxes pulmonary vessels and limits right heart hypertrophy in rats with hypoxia-induced pulmonary hypertension.

BACKGROUND: Pulmonary vascular diseases are increasingly recognised as important clinical conditions. Pulmonary hypertension associated with a range of aetiologies is difficult to treat and associated with progressive morbidity and mortality. Current therapies for pulmonary hypertension include phosphodiesterase type 5 inhibitors, endothelin receptor antagonists, or prostacyclin mimetics. However, none of these provide a cure and the clinical benefits of these drugs individually decline over time. There is, therefore, an urgent need to identify new treatment strategies for pulmonary hypertension. METHODOLOGY/PRINCIPAL FINDINGS: Here we show that the PPARbeta/delta agonist GW0742 induces vasorelaxation in systemic and pulmonary vessels. Using tissue from genetically modified mice, we show that the dilator effects of GW0742 are independent of the target receptor PPARbeta/delta or cell surface prostacyclin (IP) receptors. In aortic tissue, vascular relaxant effects of GW0742 were not associated with increases in cGMP, cAMP or hyperpolarisation, but were attributed to inhibition of RhoA activity. In a rat model of hypoxia-induced pulmonary hypertension, daily oral dosing of animals with GW0742 (30 mg/kg) for 3 weeks significantly reduced the associated right heart hypertrophy and right ventricular systolic pressure. GW0742 had no effect on vascular remodelling induced by hypoxia in this model. CONCLUSIONS/SIGNIFICANCE: These observations are the first to show a therapeutic benefit of 'PPARbeta/delta' agonists in experimental pulmonary arterial hypertension and provide pre-clinical evidence to favour clinical trials in man.

Gestió de la informació d'interès d'un treballador autònom

L’objectiu principal d’aquest treball és fer un programa que permeti portar la informació que un treballador autònom li interessa, d’acord amb les seves necessitats. En el nostre cas es tracta d’un centre d’estètica, que té més d’una seu, el qual li interessa portar tota la cartilla de clients, centres associats, proveïdors ... a més de poder fer les factures corresponents als centres associats, poder calcular en el moment que en el treballador li interessi, els ingressos realitzats durant un període de temps determinat i poder portar una agenda actualitzada dels dos centres, on es mostren totes les visites que hi ha en un dia. Per tal de realitzar el programa, s’han portat a terme mitjançant dos aplicacions, i connectant-los en una base de dades. Per una banda tenim una aplicació implementada amb C++, per l’altra, una pàgina web amb PHP, finalment com a sistema gestor de base de dades utilitzem el MySQL Server. El programa fet amb C++, consta de tota la part d’entrada i/o modificacions de dades, en aquesta part només hi pot accedir el treballador autònom, ja que és la única persona que pot fer aquesta feina. En la pàgina web, hi pot accedir qualsevol persona que tingui un nom d’usuari i una contrasenya. A través de la web es pot fer qualsevol tipus de consulta, fer tot el control de les agendes, portar a terme tot el tema de facturació i ingressos, i com a excepció l’entrada de dades de clients, ja que s’ha de poder realitzar en qualsevol moment i lloc. Per acabar, tenim la necessitat de tenir un servidor, aquest ha d’estar format, mínim, per la base de dades. Com que l’aplicació amb C++ i la base de dades han d’estar ubicades al mateix lloc. A més, necessitem un servidor web per tal de tenir la nostra pàgina a la xarxa, per aconseguir això, utilitzem un programa anomenat DynDNS, que es fa servir per a convertir una IP dinàmica en una IP estàtica i d’aquesta manera convertir un ordinador qualsevol amb un servidor web.