Pages
-
-
Context-aware recommender systems for authors of e-learning content
-
Ivana Bosnić The goal of this thesis is to research context dimensions and characteristics of context-aware recommender systems designed for authoring of e-learning content. Building on characteristics of available content models, the context model is proposed, comprising context dimensions from (i) learning management system and course context, (ii) author’s profile and (iii) context inferred from already used learning objects, which includes their structure, pedagogical roles, domain topics and...
-
-
Context-based detection of sensitive data in structured datasets
-
Vjeko Kužina The digital landscape’s growth has underscored the importance of addressing data privacy and security challenges. One such challenge, the detection of sensitive data in structured datasets, has become increasingly relevant due to regulatory frameworks like the GDPR in the EU. While the topic of sensitive data detection in unstructured data has been extensively researched, the problem of sensitive data detection in structured data has not been adequately addressed in related work. This...
-
-
Contribution to differential electromyographic diagnostics of low back pain and radiculopathy
-
Saša Ostojić Low back pain is one of the leading causes of absence from work. Diagnosing low back pain patients or mere differentiation from healthy subjects is therefore an interesting problem. One way to diagnose comes from properties of surface EMG signals to reflect complex changes occurring within skeletal muscles during fatiguing static contractions. This doctoral work introduces differentiation of low back pain patients with radiculopathy from chronic nonspecific low back pain patients and from...
-
-
Control of autonomous surface marine vessels for underwater vehicle localization using single range acoustic measurements
-
Filip Mandić Mobile beacon vehicles are used as a navigational aid for autonomous underwater vehicles when performing navigation using single range measurements. They remove the constraints imposed on the underwater vehicle trajectory by executing trajectory that provides informative range measurements. In thesis, a novel control algorithm for the beacon vehicle which ensures observability of the underwater vehicle's navigation filter is presented. The algorithm was tested in real--life environment and...
-
-
Control system for an unmanned aerial manipulator interacting with the environment based on a generalized model
-
Marko Car This thesis develops a control framework for a generalized unmanned aerial vehicle (UAV), enabling full 6 DoF control of the vehicle, as well as stable execution of contact based tasks, such as environment manipulation, contact based inspection or transport of fragile payload. The generalized UAV model developed within this thesis considers three different control modalities, namely the classical rotor thrust amplitude, rotor thrust direction, and centroid variation actuators. A unified...
-
-
Convex polar second-order Taylor approximation of AC optimal power flows
-
Karlo Šepetanc The dissertation aims to improve numerical tractability of two optimization problem categories: transmission system AC optimal power flows as stand-alone optimization and bilevel optimization problems with AC transmission network constrained market clearing in the lower level. The developed optimal power flow formulation is based on a convex approximation of Taylor expansion, as opposed to nonconvex exact formulations. Convex form is relevant for power system operation models as modern...
-
-
Convolutional neural networks for illumination estimation in complex illumination environments
-
Ilija Domislović The Human Visual System allows us to perceive the world around us. Illumination has a significant effect on an object's color, but humans perceive object color as relatively constant regardless of illumination color. This ability is called color constancy. Unlike humans, cameras are unable to do this automatically, resulting in unnatural-looking images. Methods that emulate the Human Visual System have been proposed. Research has shown that learning-based methods achieve the best results....
-
-
Coordinated optimal control of wind farm active power
-
Vedrana Spudić With large penetration of wind energy in electricity grids, a requirement to control the wind farm active power arises. To fulfill the demands set in the new grid codes wind farms must have an ability to track the power reference demanded by the transmission system operator. The research in this thesis focuses on the design of an optimal wind farm controller that enables tracking of the power reference, while achieving reduction of the wind turbine structural loads. Such control design...
-
-
Crpljenje i pretraživanje tekstnih informacija temeljem grafova događaja
-
Goran Glavaš Tekstni izvori koji opisuju događaje iz stvarnoga svijeta (novinski članci) sve su brojniji, a informacijske potrebe korisnika koje se tiču događaja sve su izraženije. Stoga su postupci za automatizirano crpljenje i pretraživanje informacija o događajima sve potrebniji. U okviru disertacije predstavljen je model grafa događaja kao strukture koja sadrži sve bitne informacijske aspekte događaja iz stvarnog svijeta. Vrhovi grafa događaja predstavljaju pojedinačna spominjanja...
-
-
Daljinski rad s programskim sustavom primjenom pokretnih agenata
-
Igor Ljubi Ovaj rad bavi se problematikom upravljanja programskim sustavom kada se on nalazi na nekom čvoru u telekomunikacijskoj mreži. Prikazani su koncepti isporuke, upravljanja i održavanja programskih sustava. Istražene su i prikazane prednosti uporabe modela pokretnih programskih agenata nad tradicionalnim načinima pristupa udaljenim objektima ili čvorovima. Temeljem provedene analize različitih implementacija programskih agenata, odabran je sustav Grasshopper. Nakon obavljenih...
-
-
Decentralized mission planning for heterogeneous robotic teams based on hierarchical task representation
-
Barbara Arbanas Pascoal Ferreira This thesis focuses on the planning and coordination of cooperative missions for heterogeneous MRS. This complex problem consists of mission decomposition selection (the question what do we do?), task allocation (the question who does what?), and task scheduling (the question how do we arrange tasks in time?), which are often summarized under the common term mission (task) planning. Mission planning can be viewed as an optimization problem that attempts to find the most appropriate way to...
-
-
Decentralizirani višeagentski sustav za interakciju sa zajednicom pčela zasnovanu na algoritmu konsenzusa
-
Karlo Griparić Evolucijski optimirani procesi odlučivanja i koordinacije između jedinki mnogih bioloških vrsta dovode do njihovog efikasnog djelovanja u cilju preživljavanja i reprodukcije. Zbog univerzalnosti pojave kompleksnog ponašanja, istraživanje algoritama i modela ponašanja jedinki u biološkim zajednicama predstavlja veliki potencijal za primjenu u višeagentskim sustavima. U disertaciji je opisan mehatronički sustav za interakciju sa zajednicom pčela, sastavljen od umreženih statičkih...
Pages
