1. Hierarchical learning of robotic contact policiesMihael Simonič, Aleš Ude, Bojan Nemec, 2023, izvirni znanstveni članek Povzetek: The paper addresses the issue of learning tasks where a robot maintains permanent contact with the environment. We propose a new methodology based on a hierarchical learning scheme coupled with task representation through directed graphs. These graphs are constituted of nodes and branches that correspond to the states and robotic actions, respectively. The upper level of the hierarchy essentially operates as a decision-making algorithm. It leverages reinforcement learning (RL) techniques to facilitate optimal decision-making. The actions are generated by a constraint-space following (CSF) controller that autonomously identifies feasible directions for motion. The controller generates robot motion by adjusting its stiffness in the direction defined by the Frenet–Serret frame, which is aligned with the robot path. The proposed framework was experimentally verified through a series of challenging robotic tasks such as maze learning, door opening, learning to shift the manual car gear, and learning car license plate light assembly by disassembly. Ključne besede: autonomous robot learning, learning, experience, compliance and impedance contro Objavljeno v DiRROS: 21.09.2023; Ogledov: 14; Prenosov: 7
Celotno besedilo (1,73 MB) Gradivo ima več datotek! Več... |
2. Algorithm instance footprint : separating easily solvable and challenging problem instancesAna Nikolikj, Sašo Džeroski, Mario Andrés Muñoz, Carola Doerr, Peter Korošec, Tome Eftimov, 2023, objavljeni znanstveni prispevek na konferenci Ključne besede: black-box optimization, algorithms, problem instances, machine learning Objavljeno v DiRROS: 15.09.2023; Ogledov: 29; Prenosov: 19
Celotno besedilo (2,03 MB) Gradivo ima več datotek! Več... |
3. Assessing the generalizability of a performance predictive modelAna Nikolikj, Gjorgjina Cenikj, Gordana Ispirova, Diederick Vermetten, Ryan Dieter Lang, Andries Petrus Engelbrecht, Carola Doerr, Peter Korošec, Tome Eftimov, 2023, objavljeni znanstveni prispevek na konferenci Ključne besede: algorithms, predictive models, machine learning Objavljeno v DiRROS: 15.09.2023; Ogledov: 36; Prenosov: 23
Celotno besedilo (935,67 KB) Gradivo ima več datotek! Več... |
4. |
5. Recombination of oxygen atoms on the surface of oxidized polycrystalline nickel—temperature and pressure dependencesDomen Paul, Miran Mozetič, Rok Zaplotnik, Jernej Ekar, Alenka Vesel, Gregor Primc, Denis Đonlagić, 2023, izvirni znanstveni članek Povzetek: The recombination of neutral oxygen atoms in the ground state on the oxidized nickel samples was studied experimentally in the range of pressures where the maximum density occurs in weakly ionized low-pressure oxygen plasma, i.e. between 40 and 200 Pa. The recombination coefficient was determined in the flowing afterglow. The source of oxygen atoms was plasma sustained in a quartz tube of inner diameter 4.7 mm by a microwave discharge in the surfatron mode. The recombination coefficient was determined in the afterglow chamber, which was a Pyrex tube with an inner diameter of 36 mm. The density of oxygen atoms in the afterglow chamber was varied by adjusting the discharge power, the gas flow, the pressure, and the position of a recombinator. Such flexibility of the experimental system enabled adjustment of the temperature of the oxidized nickel samples independently from the O-atom density in its vicinity or other parameters. The density of oxygen atoms in the afterglow chamber at various system parameters was determined by the Šorli method, which is reliable, and has an accuracy of about 20%. The recombination coefficient was determined by calorimetry. The coefficient was inversely proportional to the square root of the pressure and exponentially to the sample temperature. Systematic measurements performed at various pressures and temperatures enabled empirical formula, which were explained qualitatively by recombination kinetics.
Ključne besede: heterogenous surface recombination, recombination coefficient, nickel, nickel oxide, temperature Objavljeno v DiRROS: 30.08.2023; Ogledov: 67; Prenosov: 19
Celotno besedilo (2,44 MB) Gradivo ima več datotek! Več... |
6. DynamoRep : trajectory-based population dynamics for classification of black-box optimization problemsGjorgjina Cenikj, Gašper Petelin, Carola Doerr, Peter Korošec, Tome Eftimov, 2023, objavljeni znanstveni prispevek na konferenci Ključne besede: black-box single-objective optimization, optimization problem classification, problem representation, meta-learning Objavljeno v DiRROS: 30.08.2023; Ogledov: 77; Prenosov: 67
Celotno besedilo (650,13 KB) Gradivo ima več datotek! Več... |
7. Loss of oxygen atoms on well-oxidized cobalt by heterogeneous surface recombinationDomen Paul, Miran Mozetič, Rok Zaplotnik, Jernej Ekar, Alenka Vesel, Gregor Primc, Denis Đonlagić, 2023, izvirni znanstveni članek Povzetek: Calorimetry is a commonly used method in plasma characterization, but the accuracy of the method is tied to the accuracy of the recombination coefficient, which in turn depends on a number of surface effects. Surface effects also govern the kinetics in advanced methods such as atomic layer oxidation of inorganic materials and functionalization of organic materials. The flux of the reactive oxygen atoms for the controlled oxidation of such materials depends on the recombination coefficient of materials placed into the reaction chamber, which in turn depends on the surface morphology, temperature, and pressure in the processing chamber. The recombination coefficient of a well-oxidized cobalt surface was studied systematically in a range of temperatures from 300 to 800 K and pressures from 40 to 200 Pa. The coefficient increased monotonously with decreasing pressure and increasing temperature. The lowest value was about 0.05, and the highest was about 0.30. These values were measured for cobalt foils previously oxidized with oxygen plasma at the temperature of 1300 K. The oxidation caused a rich morphology with an average roughness as deduced from atomic force images of 0.9 µm. The results were compared with literature data, and the discrepancy between results reported by different authors was explained by taking into account the peculiarities of their experimental conditions. Objavljeno v DiRROS: 25.08.2023; Ogledov: 62; Prenosov: 20
Celotno besedilo (3,59 MB) Gradivo ima več datotek! Več... |
8. Feasibility of remote blood pressure estimation via narrow-band multi-wavelength pulse transit timeGašper Slapničar, Wenjin Wang, Mitja Luštrek, 2023, izvirni znanstveni članek Povzetek: Contact-free sensing gained much traction in the past decade. While remote monitoring of some parameters (heart rate) is approaching clinical levels of precision, others remain challenging (blood pressure). We investigated the feasibility of estimating blood pressure (BP) via pulse transit time (PTT) in a novel remote single-site manner, using a modified RGB camera. A narrow-band triple band-pass filter allowed us to measure the PTT between different skin layers, harvesting information from green and near-infrared wavelengths. The filter minimizes the inter-channel influence and band overlap, however some overlap remains within the filter bands. We further resolve this using a color-channel model and a novel channel-separation method. Using the proposed setup and algorithm, we obtained multi-wavelength (MW) PTTs in an experiment inducing BP changes to 9 subjects. The results showed good absolute Pearson’s correlation coefficient between both MW PTT and systolic BP (R = 0.61, p = 0.08) as well as diastolic BP (R = 0.54, p = 0.05), pointing to feasibility of the proposed novel remote MW BP estimation via PTT. This was further confirmed in a leave-one-subject-out experiment, where a simple Random Forest regression model achieved mean absolute errors of 3.59 and 2.63 mmHg for systolic and diastolic BP respectively.
Objavljeno v DiRROS: 27.07.2023; Ogledov: 119; Prenosov: 38
Celotno besedilo (1,06 MB) Gradivo ima več datotek! Več... |
9. The influence of catechols on the magnetization of iron oxide nanoparticlesStanislav Čampelj, Matic Pobrežnik, Tomas Landovsky, Janez Kovač, Layla Martin-Samos, Věra Hamplová, Darja Lisjak, 2023, izvirni znanstveni članek Ključne besede: magnetni nanodelci, kateholi, adsorpcija, magnetne lastnosti, XPS, DFT Objavljeno v DiRROS: 12.06.2023; Ogledov: 118; Prenosov: 52
Celotno besedilo (7,50 MB) Gradivo ima več datotek! Več... |
10. |