Drago Torkar

Research areas
  • Computer vision
  • Pattern recognition
  • Image processing
  • Machine vision

  • Projects
  • Distributed Artificial Intelligent Systems (2021-2024)
  • Computer structures and systems (2019-2024)
  • Intelligent Secure Trustable Things (2020-2023)
  • Supporting Active Ageing through Multimodal coaching (2017-2021)
  • Advanced recognition (2020)
  • Machine vision quality control of molded plastic parts (2019-2020)
  • Resource Efficient Food and dRink for the Entire Supply cHain (2015-2019)
  • iNet - The Impact of Net Height in Table Tennis (2017-2018)
  • Computer structures and systems (2015-2018)
  • Computer structures and systems (2009-2014)
  • Mobile application for food ingredients informing (2009-2010)
  • Development and implementation of a new PIM binder system using advanced methods (2008-2010)
  • The role of Luka Koper in logistic support of the Slovenian Armed Forces and allies (2006-2008)
  • Adaptive Robots for Flexible Manufacturing Systems (2005-2008)

  •  
    Adaptive Robots for Flexible Manufacturing Systems[X]
    • acronym: ARFLEX
    • type: 6FP IST
    • duration: 2005-2008
    • content:
      The project main objectives are to apply the most advanced control technologies (control theory, sensory devices, electronic embedded systems, and, in general, ICT) to radically innovate a class of products – industrial robots – where these technologies did not yet find full applications (the demanding missions have mostly been satisfied by heavy, costly, difficult to reconfigure, mechanical solutions).
      The chosen class of products is such that the success of the project will have important impacts not only in that specific class, but on all the manufacturing systems (via its diffusion and substitution potentialities).
      The scenario at the base of the project depicts a future of strong changes in many engineering products with an intimate integration of mechanical engineering with electronics, "embedding" into the design the most advanced sw/hw technologies, new intelligent sensors, complex non-linear control theory.
      The challenging objective is to overcome the mechanical limitations by embedding the intelligent use of new advanced control system theories supported by the low cost, easy to integrate, electronic and sensors devices.
      The robot mission requirements will be met by adding a continuous monitoring of the end tool positioning, feedback for real time control corrections, thus relaxing the requirements on the mechanical parts (accepting limited stiffness in the arms, large imprecision in the coupling joints, and, in general, non-linearity and friction effects).
      It will have positive impact on the mission requirements (end tool speed and precision of positioning) and on the cost reduction of the mechanical robot components.
      The project aim - to increase flexibility and adaptability, reduce cost and increase the field of applications in the job floor - will be obtained by developing the needed know how and experimental evidence for new generations of high performing embedded control systems for industrial robots in different application contexts.

  • Computing structures and systems (2004-2008)
  • Upgrade of light armoured wheeled vehicles VALUK 6x6 (2006-2007)
  • Vision interface for ABB robots (2006)
  • Secure data storage unit based on new ferroelectric semiconductor memory devices (2004-2006)
  • 2D and 3D digital map system for land, aerial and sea orientation (2004-2005)
  • Modeling rheological properties of ceramic-paraffine suspensions for low-pressure injection-moulding (2003-2004)
  • Computing structures and systems (1999-2003)
  • High performance evolutionary techniques in hardware-software codesign (1999-2001)
  • PEMCAS: personal monitor for capability supervision (1997-1998)
  • Architectural synthesis of computer systems with testability issues (1995-1998)
  • Publications
    Journal articles
  • Biomarkers of pre-existing risk of torsade de pointes under Sotalol treatment, J. ELECTROCARDIOL. (2020)
  • Dynamic features of cardiac vector as alternative markers of drug-induced spatial dispersion, J PHARMACOL TOX MET (2020)
  • Knee stiffness and viscosity of human cadaver - Wartenberg test, BJBMS (2017)
  • Cell counting tool parameters optimization approach for electroporation efficiency determination of attached cells in phase contrast images, J.Microsc. (2011)
  • Identification of radon anomalies in soil gas using decision trees and neural networks, Nukleonika (2010)
  • Application of artificial neural networks in simulating radon levels in soil gas, ChemGe (2010)
  • Visual control of an industrial robot manipulator: Accuracy Estimation, Stroj. vestn. (2009)
  • Robot Vision Accuracy Estimation, Elektroteh. vest. (2009)
  • Apparent viscosity prediction of alumina-paraffin suspensions using artificial neural networks, J. mater. process. technol. (2008)
  • Evaluation of accuracy in a 3D reconstruction system, WSEAS Trans. Syst. Control (2007)
  • Control of a robotic manipulator by visual and speech information, Eng. rev. (2002)
  • An uncalibrated robot system for reaching and grasping objects in unpredictable physical environment, J. Electr. Eng. (2001)
  • GPS positioning and digital map processing in 2D and 3D terrain environment, J. Comput. Inf. Technol. (1994)
  • Computer processing of hard-copy maps and their application in satelite positioning, J. Commun. (1994)
  • Reflection on light distribution measurement, Sens. Rev. (1992)
  • Conference papers
  • Food Waste Ontology: A Formal Description of Knowledge from the Domain of Food Waste, BFNDMA 2019 at IEEE BigData 2019
  • Analysis of an Electrocardiographic Multilead System using Artificial Neural Networks - A Study of the Dispersion during Premature Ventricular Stimulation, Biosignals 2016
  • Accuracy of a 3D reconstruction system, ISPRA 2007
  • Robot TCP positioning with vision : accuracy estimation of a robot visual control system, ICINCO 2007
  • Estimation of accuracy for robot vision control, SHR2007
  • Radon in soil gas : application of neuron networks to identify anomalies caused by earthquakes, Hazards 2006
  • Dynamic viscosity prediction of alumina-paraffin suspensions using artificial neural networks, AERC 2005
  • The use of numerical simulation for the study of low-pressure injection moulding of alumina-paraffin suspensions, AERC 2003
  • Books & Chapters in Books
  • Optimal lead selection for evaluation ventricular premature beats using machine learning approach, (2017)
  • Radon as Earthquake Precursor - Methods for Detecting Anomalies, InTech - Open Access Publisher (2011)