Digitalni repozitorij raziskovalnih organizacij Slovenije

Iskanje po repozitoriju
A+ | A- | Pomoč | SLO | ENG

Iskalni niz: išči po
išči po
išči po
išči po

Možnosti:
  Ponastavi


Iskalni niz: "ključne besede" (reinforcement) .

1 - 10 / 14
Na začetekNa prejšnjo stran12Na naslednjo stranNa konec
1.
2.
3.
Multi-user task offloading for mobile edge computing based on reinforcement learning
Jembu Mohanram Nandhini, Kaliaperumal Saravanan, Kesavan Anuratha, Sankar Uma, 2025, izvirni znanstveni članek

Povzetek: Mobile Edge computing (MEC) enables network functions and control programmable and operates key constituents of social networks in terms of increasing user’s support on devices to carry out compute. It requires traffic offloading and task scheduling to improve the storage and fast computing. In this paper, a novel method, including data driven traffic modeling enabled by a Reinforcement learning algorithm (RLTOA), is proposed for offloading traffic and improving the computing speed and minimizing the application latency of the social network. The result of the proposed data driven modeling is compared with existing methods and validate how the data driven traffic modeling for providing the computation offloading service in terms of energy budget and the mobile drop and execution of edge server. The presented computation offloading, and energy management solutions can provide valuable perceptions for practical applications of MEC. Extensive numerical findings are presented to endorse the efficacy of RLTOA and display the effect of the social network requirement.
Ključne besede: MEC, reinforcement learning, traffic offloading, task scheduling
Objavljeno v DiRROS: 18.06.2026; Ogledov: 199; Prenosov: 292
.pdf Celotno besedilo (2,09 MB)
Gradivo ima več datotek! Več...

4.
A survey on imitation learning for contact-rich tasks in robotics
Toshiaki Tsuji, Yasuhiro Kato, Gokhan Solak, Heng Zhang, Tadej Petrič, Francesco Nori, Arash Ajoudani, 2026, pregledni znanstveni članek

Povzetek: This paper comprehensively surveys research trends in imitation learning (IL) for contact-rich robotic tasks. Contact-rich tasks, which require complex physical interactions with the environment, represent a central challenge in robotics due to their nonlinear dynamics and sensitivity to small positional deviations. The paper examines demonstration collection methodologies, including teaching methods and sensory modalities crucial for capturing subtle interaction dynamics. We then analyze IL approaches, highlighting their applications to contact-rich manipulation. Recent advances in multimodal learning and foundation models have significantly enhanced performance in complex contact tasks across industrial, household, and healthcare domains. Through systematic organization of current research and identification of challenges, this survey provides a foundation for future advancements in contact-rich robotic manipulation.
Ključne besede: imitation learning, contact-rich robotic tasks, reinforcement learning, impedance control
Objavljeno v DiRROS: 20.03.2026; Ogledov: 1409; Prenosov: 377
.pdf Celotno besedilo (1,57 MB)
Gradivo ima več datotek! Več...

5.
Hunter-engaged monitoring of the Eurasian lynx during the reinforcement process
Urša Fležar, Miha Krofel, Matej Bartol, Rok Černe, Lan Hočevar, Aleksander Trajbarič, Tilen Hvala, Maruša Prostor, Matija Stergar, Malin Aronsson, 2025, izvirni znanstveni članek

Povzetek: Collaborative wildlife monitoring programs involving citizen scientists are an efficient approach for surveying large areas. In Europe, hunters play an important role in wildlife monitoring and act as crucial stakeholders in large carnivore conservation. The Eurasian lynx Lynx lynx, an elusive felid, is a species of conservation concern in Europe. In Slovenia, lynx was exterminated and later reintroduced in 1973, but the population has declined during the past decades. A reinforcement program was initiated in 2017, translocating lynx from the Carpathian population to improve the status of the critically endangered Dinaric population. The reinforcement was coupled with an intensive monitoring program, involving local hunters as key participants. In this study, we show how the collaboration between wildlife managers, researchers and hunters resulted in a robust assessment of the lynx population at a national level for a period of five years. Questionnaires distributed to hunting clubs and chance observations were used to define the expected lynx distribution, and guide the extent of systematic camera trapping surveys, involving between 63 and 101 hunters each year. In southern Slovenia, the core of the lynx population, lynx density doubled during the reinforcement period (from 0.66 to 1.30 lynx/100 km2). In north-western Slovenia where a stepping-stone population in the Alps was established in 2021, the number of lynx increased to seven. Furthermore, all three translocated females reproduced, which represents the first confirmed lynx reproduction in the Slovenian Alps in over 150 years. We discuss the motivation behind the hunters' contribution to the data collection process and the implications of this collaboration. We highlight the importance of maintaining the collaboration and their support for lynx conservation. This study serves as an example for large-scale collaborative monitoring of a recovering population undergoing intensive conservation measures with promising results, involving crucial stakeholders as citizen scientists.
Ključne besede: camera trapping, citizen science, density, reinforcement, spatial capture recapture
Objavljeno v DiRROS: 26.01.2026; Ogledov: 574; Prenosov: 417
.pdf Celotno besedilo (1,65 MB)
Gradivo ima več datotek! Več...

6.
Preliminary assessment of alkali-resistant flax based meshes for reinforcing cementitious composites
Jaka Gašper Pečnik, Laetitia Sarah Jennifer Marrot, Marica Mikuljan, Tania Langella, Matthew Schwarzkopf, 2025, izvirni znanstveni članek

Povzetek: The production of textile-reinforced concrete (TRC) requires less materials and energy in comparison with conventional concrete reinforced with steel rebars, which draws some solutions towards the production of net zero concrete that the Cement and Concrete Industry sector should reach by 2050. To go one step further, this study investigates the development of flax based meshes as basic components for a reinforcement grid in cementitious materials. Flax strands and hybrid strands (combination of flax strands and glass or basalt rovings) were impregnated with an epoxy resin to form meshes. The physical and mechanical properties of the flax based meshes were assessed and the alkaline resistance of these reinforcing materials was evaluated to determine their durability in a cementitious matrix. At isoweight of reinforcement, the flax-based meshes demonstrated the best performance in terms of specific modulus and specific strength compared to the hybrid meshes. The hybrid meshes from the control batch displayed specific properties in the same range, whether they were constituted of AR-glass or basalt. However, the use of alkali-resistant glass rovings strongly mitigated the degradation of the mechanical properties of the hybrids meshes by making them less affected by the alkaline environment, among all the reinforcement meshes. In contrast, hybrid meshes with basalt experienced an extensive reduction in tensile strength and strain after exposure to alkaline environment, due to the corrosion of the basalt fibres. Pull-out tests revealed maximum bond strengths for the flax based meshes embedded in a high-performance concrete matrix.
Ključne besede: alkali, cement, flax, reinforcement
Objavljeno v DiRROS: 27.08.2025; Ogledov: 895; Prenosov: 559
.pdf Celotno besedilo (4,01 MB)
Gradivo ima več datotek! Več...

7.
8.
9.
Use of marginal and recycled fills for GRS structures
Stanislav Lenart, Siva Ram Karumanchi, 2024, objavljeni znanstveni prispevek na konferenci

Povzetek: Marginal and recycled fills are increasingly being utilized in geosynthetic reinforced soil (GRS) structures like retaining walls and bridge abutments due to their cost- effectiveness and sustainability. Marginal fills, including locally available soils with less desirable engineering properties, can be used when properly engineered with geosynthetics to enhance strength and stability. Recycled fills, such as recycled concrete, asphalt, various ashes have been proved already as an eco-friendly alternative to conventional aggregates. However, these practices involve thorough characterization and testing of marginal and recycled fills to ensure their suitability for specific applications. Advanced geotechnical analyses, including laboratory tests and numerical modeling, help determine the optimal blend of materials and reinforcement for achieving desired performance criteria. Experiences with the use of residues from deinking paper industry and river debris as backfill material are presented. Time effect on the compaction and deformation characteristics as well as the impact of high basicity of backfill material have been considered.
Ključne besede: soil reinforcement, reinforced fill structures, GRS abutments, residues from paper industry, river debris
Objavljeno v DiRROS: 21.01.2025; Ogledov: 1283; Prenosov: 667
.pdf Celotno besedilo (581,46 KB)
Gradivo ima več datotek! Več...

10.
Iskanje izvedeno v 0.21 sek.
Na vrh