Optimal Piecewise Linear Function Approximation for GPU-based Applications

Many computer vision and human-computer interaction applications developed in recent years need evaluating complex and continuous mathematical functions as an essential step toward proper operation. However, rigorous evaluation of this kind of functions often implies a very high computational cost, unacceptable in real-time applications. To alleviate this problem, functions are commonly approximated by simpler piecewise-polynomial representations. Following this idea, we propose a novel, efficient, and practical technique to evaluate complex and continuous functions using a nearly optimal design of two types of piecewise linear approximations in the case of a large budget of evaluation subintervals. To this end, we develop a thorough error analysis that yields asymptotically tight bounds to accurately quantify the approximation performance of both representations. It provides an improvement upon previous error estimates and allows the user to control the trade-off between the approximation error and the number of evaluation subintervals. To guarantee real-time operation, the method is suitable for, but not limited to, an efficient implementation in modern Graphics Processing Units (GPUs), where it outperforms previous alternative approaches by exploiting the fixed-function interpolation routines present in their texture units. The proposed technique is a perfect match for any application requiring the evaluation of continuous functions, we have measured in detail its quality and efficiency on several functions, and, in particular, the Gaussian function because it is extensively used in many areas of computer vision and cybernetics, and it is expensive to evaluate.

New methods for the estimation of Takagi-Sugeno model based extended Kalman filter and its applications to optimal control for nonlinear systems

This paper describes new approaches to improve the local and global approximation (matching) and modeling capability of Takagi–Sugeno (T-S) fuzzy model. The main aim is obtaining high function approximation accuracy and fast convergence. The main problem encountered is that T-S identification method cannot be applied when the membership functions are overlapped by pairs. This restricts the application of the T-S method because this type of membership function has been widely used during the last 2 decades in the stability, controller design of fuzzy systems and is popular in industrial control applications. The approach developed here can be considered as a generalized version of T-S identification method with optimized performance in approximating nonlinear functions. We propose a noniterative method through weighting of parameters approach and an iterative algorithm by applying the extended Kalman filter, based on the same idea of parameters’ weighting. We show that the Kalman filter is an effective tool in the identification of T-S fuzzy model. A fuzzy controller based linear quadratic regulator is proposed in order to show the effectiveness of the estimation method developed here in control applications. An illustrative example of an inverted pendulum is chosen to evaluate the robustness and remarkable performance of the proposed method locally and globally in comparison with the original T-S model. Simulation results indicate the potential, simplicity, and generality of the algorithm. An illustrative example is chosen to evaluate the robustness. In this paper, we prove that these algorithms converge very fast, thereby making them very practical to use.

Optimizing the Density of Holes of EWT Solar Cells Taking Into Consideration Mechanical Aspects

EWT solar cells start from drilled wafers with approximately 100 holes/cm2. These holes act as stress concentrators leading to a reduction in the mechanical strength of this type of wafers. The viability of cells with higher density of holes has been studied. To this end, sets of wafers with different density of holes have been characterized. The ring on ring test has been employed and FE models have been developed to simulate the test. The statistical evaluation permits to draw conclusions about the reduction of the strength depending on the density of holes. Moreover, the stress concentration around the holes has been studied by means of the FE method employing the sub-modeling technique. The maximum principal stress of EWT wafers with twice the density of holes of commercial ones is almost the same. However, the mutual interaction between the stress concentration effects around neighboring holes is only observed for wafers with a density of 200 holes/cm2

Optimizing rabbit production under heat stress: cage density and cage size during fattening, and breeding system and water quality on rabbit does

The aim of this work was to evaluate different management strategies to optimize rabbit production under chronic heat stress. To achieve it, three trials were conducted. In the first trial, to find the optimal cage density in tropical very dry forest condition, were measured growth performance, mortality rate, injured animals and carcass performance over an initial population of 300 cross-breed rabbits of New Zealand, California, Butterfly, Dutch and Satin, weaned at 30 days (535 ± 8 g, standard error). Treatments evaluated were: 6, 12, 18 and 24 rabbits/m2 (3, 6, 9 and 12 rabbits/cage, respectively, each cage of 0.5 m2). The maximal temperature-humidity index indicated a severe heat stress from weaning to 2.2 kg body weight (experimental time). At the end of experimental period 10, 20, 30 and 30 rabbits from the treatments of 6, 12, 18 and 24 rabbits/m2, respectively, were slaughtered and carcass performance recorded. Average daily gain and feed intake decreased by 0.31 ± 0.070 and 1.20 ± 0.25 g, respectively, per each unit that the density increased at the beginning of the experiment (P = 0.001). It increased the length of the fattening period by 0.91 ± 0.16 d (P = 0.001) per each unit of increment of density. However, rabbit production (kg/m2) increased linear and quadratically with the density (P < 0.008). Animals housed at the highest density compared to the lower one tended to show a higher incidence of ringworm (68.9 vs 39.4%; P = 0.075), injured animals (16.8 vs 3.03%; P = 0.12) and mortality (20.5 vs 9.63%; P = 0.043). The proportion of scapular fat (P = 0.042) increased linearly with increasing levels of density. Increasing density reduced linearly dorsal length (P = 0.001), and reduced linear and quadratically drip loss percentage (P = 0.097 and 0.018, respectively). In the second trial, 46 nulliparous rabbit does (23 clipped and 23 unclipped) with a BW of 3.67 ± 0.05 kg (s.e.) were used to evaluate heat stress and circadian rhythms comparing unclipped and clipped rabbit does, and to study if a more extensive breeding system increase litters performance at weaning without impairing rabbit doe performance,. Rectal temperature, feed and water 4 intake were recorded for 24 h. Rabbit does were mated 7 d after circadian measurements, and randomly assigned to two breeding systems. Control (C): mated at 14 d after parturition + litter weaned at 35 d of age. Extensive (E): mate at 21 after parturition + litter weaned at 42 d of age. The first three cycles were evaluated concerning to rabbit doe and litter performance. Two hundred twenty eight weaned rabbits, were divided into two cage sizes: 0.5 and 0.25 m2 with same density (16 rabbit/m2) and growing performance was recorded. Farm and rectal temperatures were minimal and feed and water intake maximal during the night (P < 0.001). Unclipped rabbit does showed higher rectal temperature (P = 0.045) and lower feed intake respect to clipped does (P = 0.019) which suggest a lower heat stress in the latter. Kits weaned per litter was reduced by 33% (P=0.038) in C group. This reduction was more important in the 2nd and 3rd cycles compared to the first (P ≤ 0.054). Rabbit doe feed efficiency tended to decrease in E respect C group (P = 0.093), whereas it was impaired from the first to the third cycle by 48% (P = 0.014). Growing rabbits from the E group were heavier at weaning (by 38%. P < 0.001), showed a higher feed intake (+7.4%) and lower feed efficiency (-8.4%) throughout the fattening period (P ≤ 0.056) respect to C group. Cage size had minor influence in growing performance. In the third trial, forty five non pregnant and non lactating rabbit does (21 nulliparous and 24 multiparous) were assigned randomly to farm water and to potable water to study if a water quality improvement can affect positively rabbit doe response to heat stress during pregnancy and lactation. A transponder was implanted in each animal to record subcutaneous temperature at 07:30 and 14:30 h. Experimental period extended from pregnancy (with no lactation) to the next lactation (until day 28). Body temperature and milk production were recorded daily, and body condition, feed and water intake weekly. Water quality did not affect any trait (P ≥ 0.15). Pregnant rabbit does were classified as does that weaned (W: 47%), not weaned (NW: 44%) or those pregnant that did not deliver (NB: 9%). Body temperature and feed intake decreased during pregnancy (P ≤ 0.031), but water intake remained constant. In this period body temperature decreased with metabolic weight (P ≤ 0.009). In W and NW does, 5 from mating to birth energy and protein balance impaired (P≤0.011). Body temperature of W does tended to be the lowest (P ≤ 0.090). Pregnancy length and total number of kits born tended to be longer and higher in NW than in W does (P = 0.10 and 0.053, respectively). Kit mortality at birth and from birth to 14 d of lactation was high, being worse for NW than for W does (97 vs. 40%; P<0.001). Body temperature during lactation was maximal at day 12, and milk production increased it (P ≤ 0.025). . In conclusion, in our heat stress conditions densities higher than 18 rabbits/m2 (34 kg/m2) at the end of fattening, are not recommended despite cage size, gestation and lactation productivity impaired not only when lactation is extended and along successive reproductive cycles but also due to a reduced embryo/kit survival and finally water quality improvement did not attenuate negative effect of heat stress. RESUMEN El propósito de éste trabajo fue evaluar diferentes estrategias de manejo para optimizar la producción de conejos bajo estrés térmico. Para lo cual se desarrollaron tres experimentos. En el primer experimento, para encontrar el número óptimo de gazapos por m2 de jaula durante el cebo en condiciones de bosque muy seco tropical, se estudiaron los rendimientos durante el cebo, mortalidad, animales lesionados y rendimiento de la canal sobre una población inicial de 300 conejos mestizos de Nueva Zelanda, California, Mariposa, Holandés y Satin, destetados a los 30 días de edad (535 ± 8g, error estándar). Los tratamientos evaluados fueron: 6, 12, 18 y 24 conejos/m2 (3, 6, 9 y 12 conejos/jaula, respectivamente, en jaulas de 0.5 m2). Durante el período experimental (destete a 2.2 kg de peso vivo), se observaron valores de THI correspondientes con un estrés térmico severo (THI max. De 31 a 35). Al final del período experimental, 10, 20, 30, y 30 conejos de los tratamientos con densidades de 6, 12, 18 y 24 conejos/m2, respectivamente, fueron sacrificados y su canal fue valorada. El promedio de la ganancia diaria y el consumo de alimento disminuyeron en 0.31 ± 0.070 y 1.20 ± 0.25 g, respectivamente, por cada unidad de incremento en la densidad al inicio del experimento (P=0.001). Esto alargó el período de engorde en 0.91 ± 0.16 d (P=0.001) por cada unidad de incremento de la densidad. Sin embargo, la producción de conejos (kg/m2) aumentó lineal y cuadráticamente con la densidad (P<0.008). Los animales alojados en las mayores densidades en comparación con el resto tendieron a mostrar una mayore incidencia de tiña (68.9 vs 39.4%; P=0.075), de cantidad de animales heridos (16.8 vs 3.03%; P=0.12), así como de mortalidad (20.5 vs 9.63%; P=0.043). El aumento en la densidad aumentó linealmente la proporción de grasa escapular (P=0.042) y redujo linealmente la longitud dorsal (P=0.001), y lineal y cuadráticamente el porcentaje de pérdida por goteo (P=0.018). En el segundo experimento, 46 conejas nulliparas (23 rasuradas y 23 no rasuradas) con un peso vivo de 3.67 ± 0.05 kg (e.e.) fueron usadas para evaluar el estrés 8 térmico y los ritmos circadianos comparando conejas rasuradas o no, y estudiar si un sistema de crianza más extensivo mejora el desempeño de la camada al destete sin perjudicar la productividad de la coneja. Durante 24 h se midió la temperatura rectal, consumo de alimento y de agua. Las conejas fueron montadas 7 días después, y distribuidas en dos sistemas de crianza. El control (C): monta a 14 días posparto y destete a 35 d de edad. El extensivo (E): monta a 21 días posparto y destete a 42 d de edad. Se controló la productividad de la coneja y la camada durante los tres primeros ciclos. Doscientos veintiocho gazapos fueron distribuidos en dos tamaños de jaulas (0.5 y 0.25 m2) con la misma densidad (16 conejos/m2) y se controlaron sus rendimientos productivos. Durante la noche se observaron los valores mínimos para la temperatura ambiental y rectal, y los máximos para consumo de alimento y agua (P< 0.001). Las conejas no rasuradas mostraron mayor temperatura rectal (P=0.045) y menores valores de consumo de alimento con respecto a las conejas rasuradas (P=0.019), lo que sugiere un menor estrés térmico en las últimas. El número de gazapos destetados por camada se redujo en 33% (P=0.038) en el grupo C. Este comportamiento se acentuó en el 2do y 3er ciclo en comparación con el primero (P≤0.054). La eficiencia alimenticia de las conejas tendió a disminuir en el grupo E con respecto al grupo C (P=0.093), dicha tendencia se acentúa del primer al tercer ciclo en un 48% (P=0.014). Los gazapos en fase de crecimiento provenientes del grupo E fueron más pesados al momento del destete (en 38% P<0.001), mostrando un mayor consumo de alimento (+7.4%) y menor eficiencia alimenticia (-8.4%) a lo largo del engorde (P≤0.056) con respecto al grupo C. El tamaño de la jaula tuvo una mínima influencia en el comportamiento durante el crecimiento de éstos gazapos. En el tercer experimento, cuarenta y cinco conejas no gestantes ni lactantes (21 nulíparas y 24 multíparas) se les asignó al azar agua dos tipos de agua: común de la granja y agua potable, con el fin de estudiar si una mejora en la calidad del agua puede afectar positivamente la respuesta de la coneja al estrés térmico durante la gestación y la lactancia. Se les implantó un transponder para registrar la temperatura subcutánea a las 7:30 y a las 14:30 h. El período experimental se extendió desde la gestación (sin 9 lactancia) hasta la lactanción consecutiva (hasta los 28 días). La temperatura corporal y la producción de leche se controlaron diariamente, y la condición corporal, consumo de agua y alimento, semanalmente. La calidad del agua no afectó a ninguna variable (P≥0.15). Las conejas preñadas fueron clasificadas como conejas que destetaron (W: 47%), que no destetaron (NW:44%) o aquellas que no parieron (NB: 9%). La temperatura corporal y consumo de alimento disminuyeron durante la gestación (P≤0.031), mientras que el consumo de agua se mantuvo constante. La temperatura corporal descendió con el peso metabólico durante la gestación (P≤0.009). El balance de energía y proteína disminuyó desde la monta al parto para las conejas W y NW (P≤0.011). Durante la gestación la temperatura corporal tendió a ser menor en las conejas W (P≤0.090). La longitud de la gestación y el número total de gazapos nacidos tendieron a ser mayores en conejas NW que en conejas W (P=0.10 y 0.053, respectivamente). La mortalidad de los gazapos al parto y del parto a los 14 días de lactancia fue alta, siendo peor para las conejas NW que para las W (97 vs 40%; P<0.001). Durante la lactancia la temperatura corporal alcanzó su valor máximo para el día 12, y la producción de leche indujo un incremento en la misma (P≤0.025). En conclusión, en nuestras condiciones de estrés térmico y sin importar el tamaño de la jaula, no se recomiendan densidades mayores a 18 conejos/m2 (34 kg/m2) al final del engorde. La productividad de la gestación y la lactancia disminuyen cuando la lactancia es mayor y se suceden varios ciclos reproductivos seguidos. Esto se debe al efecto negativo del estrés térmico sobre la vitalidad y supervivencia del embrión/gazapo. La mejora de la calidad del agua atenuó el efecto negativo del estrés térmico. Las conejas más productoras parece que son aquéllas que consiguen manejar mejor el estrés térmico.

Embryogenic suspensions of adult cork oak: the first step towards mass propagation.

Abstract Protocols have been established to clone adult cork oak trees by somatic embryogenesis using semisolid medium. However, for economically viable mass propagation, embryogenic cultures in liquid medium need to be developed. In this study, suspension cultures were initiated from embryo clusters obtained by secondary embryogenesis on a gelled medium lacking plant growth regulators. After 6 days of culture, these embryo clusters generated high cell density suspensions that also contained small organized structures (embryos and embryogenic clumps). As the culture duration increased, tissue necrosis and fewer embryogenic structures were observed and the establishment of suspension cultures failed. An alternative method was found adequate for initiation of embryogenic suspensions: embryo clusters from gelled medium were briefly shaken in liquid medium and detached cells and embryogenic masses of 41?800 lm were used as inoculum. Maintenance of embryogenic suspensions was achieved using a low-density inoculum (43 mg l-1) by subculturing four embryogenic clumps of 0.8?1.2 mm per 70 ml of medium. Proliferation ability was maintained for almost 1 year through ten consecutive subcultures. The initiation and maintenance protocols first developed for a single genotype were effective when tested on 11 cork oak genotypes.

Optimal fits of diffusion constants from single-time data points of Brownian trajectories

Experimental methods based on single particle tracking (SPT) are being increasingly employed in the physical and biological sciences, where nanoscale objects are visualized with high temporal and spatial resolution. SPT can probe interactions between a particle and its environment but the price to be paid is the absence of ensemble averaging and a consequent lack of statistics. Here we address the benchmark question of how to accurately extract the diffusion constant of one single Brownian trajectory. We analyze a class of estimators based on weighted functionals of the square displacement. For a certain choice of the weight function these functionals provide the true ensemble averaged diffusion coefficient, with a precision that increases with the trajectory resolution.

A new approach to fuzzy estimation of Takagi-Sugeno model and its applications to optimal control for nonlinear systems

An efficient approach is presented to improve the local and global approximation and modelling capability of Takagi-Sugeno (T-S) fuzzy model. The main aim is obtaining high function approximation accuracy. The main problem is that T-S identification method cannot be applied when the membership functions are overlapped by pairs. This restricts the use of the T-S method because this type of membership function has been widely used during the last two decades in the stability, controller design and are popular in industrial control applications. The approach developed here can be considered as a generalized version of T-S method with optimized performance in approximating nonlinear functions. A simple approach with few computational effort, based on the well known parameters' weighting method is suggested for tuning T-S parameters to improve the choice of the performance index and minimize it. A global fuzzy controller (FC) based Linear Quadratic Regulator (LQR) is proposed in order to show the effectiveness of the estimation method developed here in control applications. Illustrative examples of an inverted pendulum and Van der Pol system are chosen to evaluate the robustness and remarkable performance of the proposed method and the high accuracy obtained in approximating nonlinear and unstable systems locally and globally in comparison with the original T-S model. Simulation results indicate the potential, simplicity and generality of the algorithm.

Low-Dimensional Modelling for optimal streaks in the blasius boundary layer

This paper is concerned with the low dimensional structure of optimal streaks in the Blasius boundary layer. Optimal streaks are well known to exhibit an approximate self-similarity, namely the streamwise velocity re-scaled with their maximum remains almost independent of both the spanwise wavenumber and the streamwise coordinate. However, the reason of this self-similar behavior is still unexplained as well as unexploited. After revisiting the structure of the streaks near the leading edge singularity, two additional approximately self-similar relations involving the velocity components and their wall normal derivatives are identified. Based on these properties, we derive a low dimensional model with two degrees of freedom. The comparison with the results obtained from the linearized boundary layer equations shows that this model is consistent and provide good approximations.

Influence of air drag on the optimal hand launching of a small, round projectile

Conditions leading to a maximum range for a small, round projectile, fired by hand, are discussed taking into account air drag and the dependence of the initial speed on the mass launched. Both the optimal angle of release for given projectile and initial speed, and the optimal radius for given density (i.e., among a bed of pebbles) are determined; an increase on the height of release is found to always decrease the angle and increase the radius. The influence of the projectile mass on the optimal manner of launching is considered. The validity of the approximations used in the analysis is discussed. Results from very simple measurements show good agreement with theory.

Role of surface trap states on two-dimensional electron gas density in InAlN/AlN/GaN heterostructures

In order to clarify the effect of charged dislocations and surface donor states on the transport mechanisms in polar AlInN/AlN/GaN heterostructures, we have studied the current-voltage characteristics of Schottky junctions fabricated on AlInN/AlN/GaN heterostructures. The reverse-bias leakage current behaviour has been interpreted with a Poole-Frenkel emission of electrons from trap states near the metal-semiconductor junction to dislocation induced states. The variation of the Schottky barrier height as a function of the AlN layer thickness has been measured and discussed, considering the role of the surface states in the formation of the two dimensional electron gas at AlN/GaN interface.

Power density of a bare electrodynamic tether generator

The maximum performance of bare electrodynamic tethers as power generating systems under OML-theory is analyzed. Results show that best performance in terms of power density is achieved by designing the tether in such a way to increase ohmic impedance with respect to plasma contact impedance, hence favoring longer and thinner tethers. In such condition the corresponding optimal value of the load impedance is seen to approach the ohmic impedance of the conducting tether. At the other extreme, when plasma contact impedance dominates (which is not optimal but can be relevant for some applications) optimum power generation is found by matching the load impedance with an effective tether-plasma contact impedance whose expression is derived.

Electronic structure of copper nitrides as a function of nitrogen content

he nitrogen content dependence of the electronic properties for copper nitride thin films with an atomic percentage of nitrogen ranging from 26 ± 2 to 33 ± 2 have been studied by means of optical (spectroscopic ellipsometry), thermoelectric (Seebeck), and electrical resistivity measurements. The optical spectra are consistent with direct optical transitions corresponding to the stoichiometric semiconductor Cu3N plus a free-carrier contribution, essentially independent of temperature, which can be tuned in accordance with the N-excess. Deviation of the N content from stoichiometry drives to significant decreases from − 5 to − 50 μV/K in the Seebeck coefficient and to large enhancements, from 10− 3 up to 10 Ω cm, in the electrical resistivity. Band structure and density of states calculations have been carried out on the basis of the density functional theory to account for the experimental results.

Integration of congestion pricing and intertemporal preference rate in social welfare function

Assessing social benefits in transport policy implementation has been studied by many researchers using theoretical or empirical measures. However, few of them measure social benefit using different discount rates including the inter-temporal preferences rate of users, the private investment discount rate and the inter-temporal preferences rate of the government. In general, the social discount rate used is the same for all social actors. Therefore, this paper aims to assess a new method by integrating different types of discount rate belonging to different social actors in order to measure the real benefits of each actor in the short, medium and long term. A dynamic simulation is provided by a strategic Land-Use and Transport Interaction (LUTI) model. The method is tested by optimizing a cordon toll scheme in Madrid considering socio- economic efficiency and environmental criteria. Based on the modified social welfare function (WF), the effects on the measure of social benefits are estimated and compared with the classical WF results as well. The results of this research could be a key issue to understanding the relationship between transport system policies and social actors' benefits distribution in a metropolitan context. The results show that the use of more suitable discount rates for each social actor had an effect on the selection and definition of optimal strategy of congestion pricing. The usefulness of the measure of congestion toll declines more quickly overtime.

Integration of congestion pricing and intertemporal preferences rate integration in social welfare function

Many researchers have used theoretical or empirical measures to assess social benefits in transport policy implementation. However, few have measured social benefits by using discount rates, including the intertemporal preference rate of users, the private investment discount rate, and the intertemporal preference rate of the government. In general, the social discount rate used is the same for all social actors. This paper aims to assess a new method by integrating different types of discount rates belonging to different social actors to measure the real benefits of each actor in the short term, medium term, and long term. A dynamic simulation is provided by a strategic land use and transport interaction model. The method was tested by optimizing a cordon toll scheme in Madrid, Spain. Socioeconomic efficiency and environmental criteria were considered. On the basis of the modified social welfare function, the effects on the measure of social benefits were estimated and compared with the classical welfare function measures. The results show that the use of more suitable discount rates for each social actor had an effect on the selection and definition of optimal strategy of congestion pricing. The usefulness of the measure of congestion toll declines more quickly over time. This result could be the key to understanding the relationship between transport system policies and the distribution of social actors? benefits in a metropolitan context.

A unified framework for linear function approximation of value functions in stochastic control

This paper contributes with a unified formulation that merges previ- ous analysis on the prediction of the performance ( value function ) of certain sequence of actions ( policy ) when an agent operates a Markov decision process with large state-space. When the states are represented by features and the value function is linearly approxi- mated, our analysis reveals a new relationship between two common cost functions used to obtain the optimal approximation. In addition, this analysis allows us to propose an efficient adaptive algorithm that provides an unbiased linear estimate. The performance of the pro- posed algorithm is illustrated by simulation, showing competitive results when compared with the state-of-the-art solutions.