FER Repository
Not a member yet
12128 research outputs found
Sort by
Accessibility of countable sets in plane embeddings of arc-like continua
No full textWe consider the problem of finding embeddings of arc-like continua in the plane for which each point in a given subset is accessible. We establish that, under certain conditions on an inverse system of arcs, there exists a plane embedding of the inverse limit for which each point of a given countable set is accessible. As an application, we show that for any Knaster continuum K, and any countable collection C of composants of K, there exists a plane embedding of K in which every point in the union of the composants in C is accessible. We also exhibit new embeddings of the Knaster bucket-handle continuum in the plane which are attractors of plane homeomorphisms, and for which the restriction of the plane homeomorphism to the attractor is conjugate to a power of the standard shift map.
Investigating the value of end-users controllability on improving network operational regimes
No full textThe continuous increase in the share of distributed energy resources (DER) impact everyday operation and create new challenges for distribution system operators (DSOs). In this paper, we investigate three novel concepts and their potential for DSOs. The hosting capacity (HC) concept is utilized to calculate the available network capacity for new connections during the planning phase and is the first concept investigated in this paper. We recognize the shortcomings of the HC concept and further analyze how applying the dynamic operating envelopes (DOE) concept creates the possibility to increase the network’s capacity. By additional control of reactive power, it is possible to determine the total flexibility potential of the distribution network in the form of P-Q flexibility region which creates the need for the existence of local flexibility markets. Through various case studies, the impact of end-user controllability and flexibility on network conditions and change of export and import limits are analyzed. The analysis results showed that the DOE concept enable higher local production in relation to HC, while the exchange power with MV network remains similar. On the other hand the P-Q flexibility regions concept can increase this exchange power by 20%
Okvir za višemodalnu podvodnu interakciju čovjeka i robota
No full textThe need for understanding our marine ecosystem increases every year, both for scientific and
economic reasons. This makes marine robotics, among other research disciplines, more and
more popular and interesting. The advancements in underwater robotics have great potential
for facilitating human tasks in unstructured and hazardous underwater environments. By addressing challenges such as limited visibility, underwater communication, navigation, and diver
safety, recent research has enabled effective collaboration between autonomous underwater vehicles (AUVs) and human divers. These efforts have the common goal of improving the safety,
reliability, and efficiency of underwater operations.
This thesis explores sonar technology and its applications, focusing on sonar simulation and
diver detection methods, as well as different underwater human-robot interaction modalities.
The work begins with the theoretical principles of underwater acoustics and sonar operation.
Building on this foundation, the thesis introduces a sonar simulator developed to model underwater acoustic effects and generate realistic synthetic sonar images. The goal of the simulator
as a tool is to accelerate the development of sonar perception methods and provide a platform
for testing in a variety underwater scenarios. Further research integrates machine learning techniques with sonar imagery for real-time diver detection. By leveraging synthetic data generation
and extensive collections of real-world data, the methods aim to enhance diver-robot interaction
and autonomous underwater vehicle (AUV) navigation. Finally, the thesis addresses underwater
human-robot interaction, emphasizing gesture-based communication as a promising approach
in diver-robot interaction. The work includes creation of a comprehensive dataset for training
and validating gesture recognition algorithms, advancing intuitive and effective communication
modalities between divers and robots.
This thesis offers a comprehensive overview of sonar technology as an underwater perception method, and connects it with the underwater human-robot interaction concept through
simulation and field experiments. By incorporating simulation tools, advancements in detection
methods, and novel interaction modalities, this research establishes the foundation for future
advancements in sonar-based technologies and diver-robot interaction.Sve veća potreba za razumijevanjem morskih ekosustava, iz znanstvenih i gospodarskih razloga, potiče razvoj i zanimanje za pomorsku robotiku. Napredak u podvodnoj robotici ima velik potencijal za podršku ljudima u zahtjevnim podvodnim okruženjima. Rješavanjem izazova poput loše vidljivosti, komunikacije, navigacije i sigurnosti ronilaca omogućena je suradnja između autonomnih podvodnih vozila (AUV) i ronilaca s ciljem povećanja sigurnosti i učinkovitosti.
Ova disertacija proučava sonarsku tehnologiju kroz simulaciju, detekciju ronilaca i podvodnu interakciju čovjek-robot. Rad počinje teorijskim osnovama podvodne akustike i principa rada sonara. Zatim se predstavlja simulator sonara koji modelira akustične efekte i generira sintetske slike sonara radi bržeg razvoja metoda percepcije i testiranja u raznim scenarijima. Dalje se istražuje primjena strojnog učenja za detekciju ronilaca u stvarnom vremenu koristeći sintetske i stvarne podatke. Ispituje se i komunikacija gestama kao prirodan način interakcije između robota i ronilaca, uz izradu skupa podataka za treniranje algoritama prepoznavanja gesta.
Disertacija povezuje percepciju sonarom s podvodnom interakcijom čovjek-robot kroz simulacije i terenska ispitivanja, postavljajući temelje za daljnji razvoj sonarskih tehnologija i suradnje s roniocima
Towards an Accessible Metaverse Experience: Evaluation of a Multiplatform Technological Heritage Museum Prototype
No full textBefore metaverse technologies become fully integrated into daily life, their accessibility must be carefully considered. To ensure equal opportunities for all users, regardless of age or disability, immersive technologies should offer seamless and intuitive interaction with virtual environments, objects, and other users. This paper presents an evaluation of the accessibility and user experience of a metaverse technological heritage museum prototype on two platforms: mobile devices and virtual reality. Through feedback from 64 participants of various ages, we define accessibility guidelines for metaverse museums and identify requirements for improving the prototype. Our findings reveal significant differences between young participants and adults in their navigation and interaction experiences across platforms. This work addresses a research gap in metaverse museum accessibility evaluation and contributes to the development of more inclusive virtual spaces by providing concrete recommendations aligned with accessibility standards
Sinteza dijagrama zračenja antenskih nizova s visokom učinkovitošću glavne latice
No full textThe design of antenna arrays forming narrow beams usually takes into account two sets of requirements. The first set shapes the beam pattern, with the aim of achieving high beam efficiency for specified sidelobe level. The second arises from implementation aspects, which includes the dynamic range ratio (DRR) of excitation coefficients and the computational complexity of the algorithm for determining the excitation coefficients. Since these requirements are contradictory, the array design involves a tradeoff. To achieve it, the dissertation considers two approaches, polynomial synthesis and numerical optimization. In the first approach, the dissertation presents three methods for the synthesis of pencil beams and one for the synthesis of narrow flat-top beams. In the second, the dissertation offers two methods for the design of pencil-beam arrays with high beam efficiency and reasonably low DRR. In the polynomial synthesis of pencil beams, the dissertation proposes Pascal polynomials, derivatives of Chebyshev polynomials, and Gegenbauer polynomials. For the polynomial synthesis of flat-top beams, the dissertation introduces modifications of Chebyshev polynomials of first kind. Finally, the dissertation proposes the control of DRR by using a global optimization framework. By using this framework, the dissertation proposes a method for the design of linear antenna arrays with specified DRR and maximum beam efficiency and a method for the design of arrays with minimum DRR and specified beam efficiency.Dizajn antenskih nizova s uskom glavnom laticom obično uzima u obzir dva skupa zahtjeva. Prvi skup oblikuje dijagram zračenja antenskog niza, s ciljem postizanja visoke učinkovitosti glavne latice za danu razinu sekundarnih latica. Drugi skup proizlazi iz implementacijskih aspekata, uključujući dinamički raspon pobudnih koeficijenata (DRR) i računsku složenost algoritma za određivanje pobudnih koeficijenata. S obzirom da su ovi zahtjevi kontradiktorni, dizajn niza predstavlja kompromis. Za postizanje kompromisa, disertacija razmatra dva pristupa: polinomnu sintezu i numeričku optimizaciju. U prvom pristupu, disertacija predstavlja tri metode za sintezu dijagrama zračenja s uskom glavnom laticom i jednu metodu za sintezu dijagrama zračenja s ravnim vrhom. U drugom pristupu, disertacija predlaže dvije metode za dizajn nizova s visokom učinkovitošću glavne latice i malim DRR-om. U polinomnoj sintezi dijagrama zračenja s uskom glavnom laticom, disertacija predlaže Pascalove polinome, derivacije Čebiševljevih polinoma i Gegenbauerove polinome. Za polinomnu sintezu dijagrama zračenja s ravnim vrhom, disertacija uvodi modifikaciju Čebiševljevih polinoma prve vrste. Konačno, disertacija predlaže kontrolu DRR-a korištenjem globalne optimizacije. Na temelju globalne optimizacije, disertacija predlaže metodu za dizajn linearnih antenskih nizova sa zadanim DRR-om i maksimalnom učinkovitošću glavne latice te metodu za dizajn linearnih antenskih nizova s minimalnim DRR-om i maksimalnom učinkovitošću glavne latice
Genetic programming with surrogate evaluation for the electric vehicle routing problem
No full textThe focus on environmental sustainability has made the Electric Vehicle Routing Problem (EVRP) an important area of research. Routing Policies (RPs) offer a simple and efficient approach to solving VRPs, providing advantages over methods like metaheuristics by quickly generating solutions. However, designing efficient RPs manually can be time-consuming. Therefore, there is a need to explore hyper-heuristic approaches, particularly Genetic Programming (GP), to automate the design of RPs. However, population-based evolutionary algorithms like GP often require a significant amount of computational resources, especially for fitness calculation. Therefore, surrogate evaluation is essential in enhancing efficiency, especially in GP, where multiple problem instances need to be solved to evaluate each chromosome. In this study, we employ surrogate models within GP to design RPs for EVRP with hard time windows. The experiments show that the RPs designed by GP with surrogate models outperform those produced by standard GP approaches while still requiring less computational time to be generated. Moreover, the RPs designed with GP using surrogate models are also smaller, and consequently, they are also more efficient and easier to interpret
Large Language Model-Driven Framework for Automated Constraint Model Generation in Configuration Problems
No full textConstraint satisfaction problems (CSPs) are widely used in domains such as product configuration, scheduling, and resource allocation. However, formulating constraint models remains a significant challenge that often requires specialized expertise in constraint programming (CP). This study introduces the Automatic Constraint Model Generator (ACMG), a novel framework that leverages fine-tuned large language models (LLMs) to automate the translation of natural language problem descriptions into formal CSP models. The ACMG employs a multi-step process involving semantic entity extraction, constraint model generation, and iterative validation using the MiniZinc solver. Our approach achieves state-of-the-art (SOTA) or near-SOTA results, demonstrating the viability of LLMs in simplifying the adoption of CP. Its key contributions include a high-quality dataset for fine-tuning, a modular architecture with specialized LLM components, and empirical validation which shows its promising results for complex configuration tasks. By bridging the gap between natural language and formal constraint models, the ACMG significantly lowers the barrier to CP, making it more accessible to non-experts while maintaining a high level of robustness for industrial applications
Online grocery shopping in Europe: consumer research on the role of innovation and local food perceptions
No full textPurpose: This study investigates the determinants of online grocery shopping adoption in Europe, focusing on the effect of innovation adoption characteristics, food choice motives and local food perceptions on attitudes and purchase intentions. Based on the significant determinants for adoption, consumer segments are identified to gain insights into the diversity in online grocery shopping behaviour.
Design/methodology/approach: A pan-European online survey was conducted in March 2024 with 2,899 respondents from Belgium, France, the UK, Spain and Croatia. The data were analysed using partial least squares structural equation modelling to examine relationships between innovation adoption, food preferences and local food perceptions. A segmentation analysis was also performed to identify distinct consumer groups.
Findings: The research identifies the innovation adoption characteristics, perceived convenience and subjective norms as crucial in shaping purchase intentions across different countries. Additionally, the study reveals insights about the positive influence of e-commerce-related food choice motives on consumer attitudes, while perceived complexity, perceived product risk and local food perception showed negative effects. Five consumer segments were identified, with digital pioneers and complexity-averse consumers showing the highest engagement potential.
Originality/value: This research provides a first multi-country assessment for the determinants of the adoption of online grocery shopping in a multi-country European context. New insights into the role of local food perceptions in online grocery shopping adoption are gathered, which can support the market access for small- and medium-sized producers
Reverse engineering of iOS mobile applications and defensive techniques
No full textReverznim inženjerstvom moguće je u značajnoj mjeri otkriti unutarnje mehanizme funkcioniranja mobilnih aplikacija, a otkrivanjem unutarnjih mehanizama potencijalno se otkrivaju i sigurnosne ranjivosti koje je moguće zlonamjerno iskoristiti. Ovaj rad u fokus stavlja iOS mobilne aplikacije te prikazuje primjer provođenja postupka reverznog inženjerstva kojim su otkrivene sigurnosne ranjivosti. U radu se opisuje postupak povišenja privilegija unutar operacijskog sustava iOS, poznatiji pod nazivom Jailbreak. Prikazuje se postupak dohvaćanja i dešifriranja mobilne aplikacije iz datotečnog sustava mobilnog uređaja iPhone te se analizira njen sadržaj i opisuju otkrivene ranjivosti. Rad detaljno opisuje Mach-O strukturu izvršne datoteke mobilne aplikacije čije je poznavanje nužno za uspješno reverzno inženjerstvo te ga uspoređuje s formatima izvršnih datoteka drugih operacijskih sustava. Napadi na iOS mobilne aplikacije temeljeni na reverznom inženjerstvu u radu se kategoriziraju u statičke i dinamičke napade te se na temelju primjera opisuju pojedinačni napadi iz svake od kategorija. Za svaki napad predlažu se mogući načini zaštite od istog, a poseban fokus se usmjerava na zaštitu u vidu obfuskacije i provjere integriteta čiji se značaj demonstrira korištenjem komercijalnih alata. Zaključno, rad naglašava važnost pravno-regulatornih aspekata opisanih postupaka kako bi se tijekom njihovog korištenja osigurala usklađenost sa zakonom te izbjegle moguće kazne.By using the reverse engineering techniques, it is possible to significantly reveal the internal mechanisms of mobile applications. The discovery of internal mechanisms potentially also reveals security vulnerabilities that can be exploited in a malicious way. This paper focuses on iOS mobile applications and describes an example of conducting the reverse engineering procedure that revealed security vulnerabilities. The paper describes the process of privilege elevation within the iOS operating system, better known as Jailbreak. It shows the process of retrieving and decrypting a mobile application from the file system of an iPhone mobile device, analyzes its content and describes the detected vulnerabilities. The paper describes the details of the Mach-O structure in executable files important for effective reverse engineering and compares it with the executable file formats from other operating systems. Reverse engineering-based attacks on iOS mobile applications are classified into static and dynamic types, with examples provided for each category. Possible defense strategies are proposed for each attack, with particular focus on protection through obfuscation and integrity checks demonstrated using commercial tools. In conclusion, the paper emphasizes the importance of legal-regulatory aspects of the described procedures to ensure compliance with the law during their use and avoid possible penalties