1,720,960 research outputs found
Evolving State Machines as Robot Controllers
No full textAutomated generation of robot controllers using an Evolutionary Algorithm(EA) has received increasing attention in the last years as it has the potentialfor a reduction in the development time of a robot. Often these EAs generateNeural Networks (NNs) as robot controllers. Using a NN for automaticallygenerating robot controllers has two important downsides: 1.) A human isnot able to fully understand the inner working of a multi-layer NN, and 2.)a NN has only limited abilities to decompose a complex task into sub tasks.Both of these downsides can be addressed by using a State Machine (SM)instead of a NN as robot controller. Therefore, this thesis introduces an EAcalled Evolving State Machines As Controllers (ESMAC). ESMAC generatesSMs instead of NNs. A SM is understandable for humans because ofits modularity and allows for task decomposition by using a state for eachsub task. Furthermore, two extensions of ESMAC are proposed: adaptiveESMAC and selector ESMAC. Adaptive ESMAC aims to automatically determinesthe number of states with which the best tness for a task canbe achieved. Selector ESMAC replaces the transitions that are used in aSM to switch between states with a NN-based switching mechanism. This switching mechanism allows mutations to make more gradual changes to aSM's behaviours, which improves the performance of the EA. The performance of ESMAC is evaluated on two robotic tasks: come-and-go and phototaxis-with-obstacles. All three variants of ESMAC showequally good performance as a NN-based EA on the evaluated tasks. Thecontrollers generated with standard ESMAC and adaptive ESMAC hardlymake any state transitions and mainly use one state. However, controllers that do use multiple states appear to be more robust to changing scenarios and in noisy environments. Selector ESMAC is able to generate SMs-based controllers that have complementing states and, therefore, shows potentialfor decomposing a task into sub tasks.Zebro projec
Asimilacija podatkov za izboljšane ocene razrešnice z wflow_sbm modelom: študija primera reke Overijsselse Vecht (Nizozemska)
No full textExtreme hydrological events have become more frequent, as evidenced by the European floods of July 2021, which affected the southern provinces of the Netherlands. The need for improved discharge predictions to be used in operational water management to avoid potential adverse effects of flooding has encouraged researchers to employ several ways to improve hydrological model estimates, including data assimilation. This thesis explores the data assimilation effects in the discharge predictions of the wflow_sbm distributed hydrological model of the Vecht river basin. Additionally, effects on other hydrological states and fluxes like subsurface flow, saturated water depth, and soil moisture were explored spatially. This work presents a methodology for applying data assimilation in a model where water is routed from the surface and subsurface. In contrast, previous studies used a model in which water is routed only via surface water. Ensemble Kalman Filter is used to update the model’s discharge predictions by assimilating external discharge observations. This methodology also explores how the data assimilation effect is influenced by the uncertainty characterization considered in the assimilation framework and other factors like the length of the assimilation window and the number of assimilation locations. A preliminary study of the rainfall data is performed to determine the uncertainties of the chosen rainfall product. A benchmark simulation scenario is then selected after the review of deterministic and ensemble model predictions. Finally, data assimilation experiments are developed after discussing the characterization of the uncertainty model. The results of the model output analysis indicate that streamflow assimilation typically has a positive effect on improving model discharge estimations. Additionally, the Ensemble Kalman Filter update effectively captures the system’s spatial state dynamics for subsurface states and fluxes, such as saturated water depth, soil moisture, etc. Two alternative experimental setups with different assimilation intervals and numbers of assimilated observations are examined concerning how this effect varies over other flow gauge locations. As demonstrated by both experiments, longer assimilation times give better results, with the assimilation effect significantly improving in the final timesteps of the assimilation frame. Furthermore, it is concluded that assimilation of observations near the outlet and interior gauges will improve discharge predictions, whereas assimilation of observations only near the outlet will only improve discharge predictions at a number of stations, typically those that are closer to the assimilation location and those where the wflow sbm model exhibits the same trend as the assimilation station. An uncertainty factor of 2.5 for the precipitation error and 0.1 for the observation error yielded the best results for both experiments. However, this study has several limitations, including assumptions of a perfect model and initial conditionsthe way the precipitation and observations error model was derived. As a result, the model gives unrealistic discharge predictions when compensating for the neglected errors. Additionally, a limited number of experiments due to the extensive computational times, attributed to the combination of the OpenDA tool with the distributed model, and the algorithm choice, does not allow the DA impact on the discharge predictions to be judged accurately. Therefore, the final section of this study provides recommendations for future research, suggesting additional experiments with longer assimilation windowsanalysis of the spatial correlation structure of precipitation, the use of more statistically reliable techniques to assess the precipitation uncertaintiesconsideration of the model parameter and initial conditions uncertaintyetc.Ekstremni hidrološki dogodki so postali vse pogostejši, kar dokazujejo evropske poplave julija 2021, ki so prizadele južne nizozemske province. Potreba po izboljšanih napovedih izpustov, ki se uporabljajo pri operativnem upravljanju voda, da bi se izognili morebitnim škodljivim učinkom poplav, je spodbudila raziskovalce k uporabi več načinov za izboljšanje ocen hidroloških modelov, vključno z asimilacijo podatkov. Diplomsko delo raziskuje učinke asimilacije podatkov pri napovedih pretoka porazdeljenega hidrološkega modela wflow_sbm porečja reke Vecht. Poleg tega so bili prostorsko raziskani učinki na druga hidrološka stanja in tokove, kot so podzemni tok, globina nasičene vode in vlažnost tal. To delo predstavlja metodologijo za uporabo asimilacije podatkov v modelu, kjer je voda usmerjena s površine in pod površino. V nasprotju s tem so prejšnje študije uporabile model, v katerem je voda speljana le po površinski vodi. Ensemble Kalmanov filter se uporablja za posodobitev napovedi izpustov modela z asimilacijo zunanjih opazovanj izpustov. Ta metodologija raziskuje tudi, kako na učinek asimilacije podatkov vpliva karakterizacija negotovosti, upoštevana v asimilacijskem okviru, in drugi dejavniki, kot sta dolžina asimilacijskega okna in število asimilacijskih lokacij. Izvede se predhodna študija podatkov o padavinah, da se določijo negotovosti izbranega produkta padavin. Po pregledu napovedi determinističnih in ansambelskih modelov se nato izbere primerjalni simulacijski scenarij. Po razpravi o karakterizaciji modela negotovosti so razviti poskusi asimilacije podatkov. Rezultati analize rezultatov modela kažejo, da ima asimilacija toka običajno pozitiven učinek na izboljšanje ocen pretoka modela. Poleg tega posodobitev filtra Ensemble Kalman učinkovito zajame dinamiko prostorskega stanja sistema za stanja in tokove pod površino, kot so globina nasičene vode, vlažnost tal itd. Dve alternativni eksperimentalni nastavitvi z različnimi intervali asimilacije in številom asimiliranih opazovanj sta preučeni glede tega, kako ta učinek razlikuje glede na druge lokacije merilnika pretoka. Kot sta dokazala oba poskusa, dajejo daljši časi asimilacije boljše rezultate, pri čemer se učinek asimilacije znatno izboljša v končnih časovnih korakih okvira asimilacije. Poleg tega je ugotovljeno, da bo asimilacija opazovanj v bližini izpusta in notranjih merilnikov izboljšala napovedi pretoka, medtem ko bo asimilacija opazovanj samo v bližini iztoka izboljšala le napovedi pretoka na številnih postajah, običajno tistih, ki so bližje lokaciji asimilacije, in tistih kjer model wflow sbm kaže enak trend kot asimilacijska postaja. Faktor negotovosti 2,5 za napako padavin in 0,1 za napako opazovanja je dal najboljše rezultate za oba poskusa. Vendar ima ta študija več omejitev, vključno s predpostavkami o popolnem modelu in začetnih pogojihnačin, kako je bil izpeljan model napak padavin in opazovanj. Posledično daje model nerealne napovedi praznjenja pri kompenzaciji zanemarjenih napak. Poleg tega omejeno število poskusov zaradi obsežnih računskih časov, pripisanih kombinaciji orodja OpenDA s porazdeljenim modelom, in izbire algoritma ne omogoča natančne ocene vpliva DA na napovedi praznjenja. Zato zadnji del te študije podaja priporočila za prihodnje raziskave in predlaga dodatne poskuse z daljšimi asimilacijskimi oknianaliza prostorske korelacijske strukture padavin, uporaba statistično zanesljivejših tehnik za ocenjevanje padavinske negotovostiupoštevanje negotovosti parametrov modela in začetnih pogojevitd
An indicator of the multiple equilibria regime of the Atlantic meridional overturning circulation
No full textRecent model results have suggested that there may be a scalar indicator ? monitoring whether the Atlantic meridional overturning circulation (MOC) is in a multiple equilibrium regime. The quantity ? is based on the net freshwater transport by the MOC into the Atlantic basin. It changes sign as soon as the steady Atlantic MOC enters the multiple equilibrium regime because of an increased freshwater input in the northern North Atlantic. This paper addresses the issue of why the sign of ? is such a good indicator for the multiple equilibrium regime. Changes in the Atlantic freshwater budget over a complete bifurcation diagram and in finite amplitude perturbation experiments are analyzed in a global ocean circulation model. The authors show that the net anomalous freshwater transport into or out of the Atlantic, resulting from the interactions of the velocity perturbations and salinity background field, is coupled to the background (steady state) state freshwater budget and hence to ?. The sign of ? precisely shows whether this net anomalous freshwater transport is stabilizing or destabilizing the MOC. Therefore, it can indicate whether the MOC is in a single or multiple equilibrium regime.<br/
Sensitivity of the Atlantic meridional overturning circulation to South Atlantic freshwater anomalies
No full textThe sensitivity of the Atlantic Meridional Overturning Circulation (AMOC) to changes in basin integrated net evaporation is highly dependent on the zonal salinity contrast at the southern border of the Atlantic. Biases in the freshwater budget strongly affect the stability of the AMOC in numerical models. The impact of these biases is investigated, by adding local anomaly patterns in the South Atlantic to the freshwater fluxes at the surface. These anomalies impact the freshwater and salt transport by the different components of the ocean circulation, in particular the basin-scale salt-advection feedback, completely changing the response of the AMOC to arbitrary perturbations. It is found that an appropriate dipole anomaly pattern at the southern border of the Atlantic Ocean can collapse the AMOC entirely even without a further hosing. The results suggest a new view on the stability of the AMOC, controlled by processes in the South Atlantic. <br/
Napovedljivost poplav na hudourniških prispevnih območjih vodotokov Rur, Niers in Swalm na Nizozemskem
No full textThe Dutch province Limburg is characterized by sloping terrain and is therefore prone to serious damages during flash floods. This was also the case for the July 2021 flood event. The study focuses on a detailed hydrological analysis and the generation of deterministic flood forecasts of the tributaries of the Meuse River in Limburg (i.e. Rur, Niers and Swalm rivers). For hydrological modeling, the wflow_sbm model was used which is a distributed hydrological model and its parameters were estimated with the Iterative Hydrography Upscaling (IHU) method and from Pedo Transfer Functions (PTFs). The horizontal hydraulic conductivity fraction (KsatHorFrac) parameter for the Rur catchment was calibrated to generate deterministic forecasts. Due to the significant difference between the observed and the simulated discharges of the Niers and Swalm catchments it was not possible to calibrate their hydrological models. Probable reasons for this difference can be the effects of ground water abstractions for pit mining and other purposes, frequent mowing management in the Niers catchment etc. Therefore forecasts were generated only for the Rur catchment by coupling the wflow model with Delft-FEWS. The forecasts generated using the DWD ICON dataset showed substantial error when compared to the observed discharge. However, for the flood event of 2021, the model predicted high flows 5-6 days ahead of the flood. There was a large overestimation of the peak for the forecast in downstream of the Rur catchment. The sensitivity of the forecast performance by changing KsatHorFrac was also analysed in the end. The findings of the study show several scopes of improvements in the wflow hydrological modeling and flood forecasting of the Meuse tributaries. Future studies based on these recommendations could aid in providing a more accurate flood prediction in this region.Območje Limburga na Nizozemskem z relativno strmo topografijo terena je med poplavami julija 2021 utrpelo veliko poplavno škodo. Magistrska naloga se osredotoča na podrobne hidrološke analize in izdelavo deterministične napovedi poplav pritokov reke Meuse na območju Limburga (prispevna območja rek Rur, Niers in Swalm). Za potrebe hidrološkega modeliranja je bil uporabljen prostorsko distribuiran hidrološki model wflow_sbm, parametri modela so bili ocenjeni z metodo Iterative Hydrography Upscaling (IHU) in s pomočjo Pedo Transfer Function (PTF). Parameter horizontalne hidravlične prevodnosti (KsatHorFrac) v hidrološkem modelu prispevnega območja reke Rur je bil umerjen in na podlagi tega so bile izdelane deterministične hidrološke napovedi. Zaradi znatnih razlik med opazovanimi in simuliranimi pretoki rek Niers in Swalm ni bilo mogoče detajlno umeriti hidroloških modelov za prispevni območji teh dveh rek. Najverjetnejši vzroki za ugotovljena razhajanja med merjenimi in modeliranimi pretoki so velike količine odvzete vode za različne rabe ter izvaje regulacijskih in vzdrževalnih del v strugi reke Niers. Zato so bile hidrološke napovedi izdelane za prispevno območje reke Rur z združeno uporabo modelov wflow in Delft-FEWS. Hidrološke napoved, ustvarjena z uporabo nabora vhodnih hidrometeoroloških podatkov iz baze DWD ICON, so se izkazale kot precej nenatančne. Navkljub slabšim rezultatom simulacij je bil model sposoben za poplavni dogodek, ki se je zgodil leta 2021, napovedal visokovodne razmere 6 dni pred dejanskim pojavom poplavnega dogodka. Simulirane vrednosti pretoka so bile precenjene predvsem v spodnjem delu prispevnega območja reke Rur. Na koncu smo analizirali tudi občutljivost hidroloških napovedi na spremenjene vrednosti parametra KsatHorFraca. Izsledki študije nakazujejo možnosti izboljšav pri hidrološkem modeliranju z uporabo programa wflow in napovedovanju poplav hudourniških pritokov reke Meuse. Nadaljnje študije, ki temeljijo na naših ugotovitvah, bi lahko zagotovile natančnejšo napoved poplav v obravnavani regiji
Going Beyond Counting First Authors in Author Co-citation Analysis
No full textThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
No full text“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
No full textWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Dispelling the Myths Behind First-author Citation Counts
No full textWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
- …
