Stefan Kozak

From e-Learning to Industry 4.0

Keywords like Flexible Learning revolution, e-Learning, Life-Long Learning, Learning Management Systems (LMS), Blended Learning, Flipped Classrooms, Learning Analytics, or Massive Open Online Courses (MOOCs) represent just few of the headlines dominating the last two decades of the education development. All of them are strongly influenced and conditioned by the Internet phenomenon. Of course, this list is still far from being complete. One could continue with numerous more recent keywords as Cloud Computing, Internet of Things, or Industry 4.0 that are again strongly related to the Internet. All these words appeared as slogans of numerous campaigns in the society development typically having a hype-cycle features. The presentation commemorates several moments of previous development, the good, the bad and the ugly. In such a context it introduces the latest initiative presented as Industry 4.0 and its importance for the automotive industry in Slovakia. Then it raises several questions concerning sustainability of the future development. Based on previous experience it concludes by the statement that the slogan Industry 4.0 is not complete. If it will not simultaneously include also massive extension to Industry. 4.University and Industry.4.Teachers activities, a sustainable development will be seriously jeopardized, at least in the Central Europe.

Advanced methods of controller design: Simulation and co-simulation on FPGA

The presented paper deals with design, co-simulation and comparison of three controller design methods (Internal Model Control, Pole-Placement, PID by Modulus Optimum) applied in the control of processes with complex dynamics. The presented and tested methods guarantee robust stability and high performance of controlled system. Proposed algorithms were successfully implemented in Matlab-Simulink using System Generator Toolbox and co-simulated on the Field Programmable Gate Array (FPGA) technology for the control of motor system model.

From PID to MPC: Control engineering methods development and applications

Process automation systems are widely known to be a crucial element in processing plants. Power plants, chemical, petrochemical, oil refining, food, automotive and many other processes in industry are vast and complex, and said to be controlled by just a handful of major players worldwide. Computer control systems and control engineering methods are at the heart of all automatically controlled processes. Motivated by practical success of modern control engineering methods in many industrial process and consumer electronics products there has been an increasing amount of work on development of new methods based on new robust, adaptive methods, numerical optimization techniques, soft computing strategies, microcomputers, decentralized control structure and hardware embedded controllers realization. The paper presents current state and development of the currently most widely used control methods, structures, and their applications to different types of automated processes. The objective of the paper is to survey and compare performance of conventional controllers with several selected advanced controllers with regard to their advantages and drawbacks in industrial applications.

Advanced methods of controller design for pneumatic servodrives

A development of advanced control methods based on robustness, optimality and prediction is an important and challenging task. In recent years, a new control approaches have emerged to replace or augment conventional control engineering methods. The paper presents: design, verification and comparison of three advanced methods (IMC, MPC, PLM) applied in the control of dynamic processes with complex dynamics. In this paper three different methods were tested and applied to the position control of pneumatic servodrives. The presented and tested methods guarantee robust stability and high performance of the pneumatic servodrives control.

Virtual Laboratory Based on Node.js Technology and Visualized in Mixed Reality Using Microsoft HoloLens.

The paper demonstrates remote control of test experiment in the virtual laboratory. This is a common problem, but another way can always be used to solve it. The paper compares several existing virtual laboratories and their possible issues at present. To develop such a new solution JavaScript technology was used on both client and server side using Node.js runtime. The modern approach is a visualization of received data in mixed reality using Microsoft HoloLens or another compatible device with Windows Mixed Reality platform.

Object recognition by effective methods and means of computer vision

The paper deals with the design and new solutions of application software with the aim to detect and recognize objects sensed by a camera. Objects of the sensed scene were determined and recognized after previous digital processing of data delivered by the camera. To this end the computer vision learning neural-based methods of feature extraction were used. The proposed application software may be used in various applications where the tracking of objects and understanding of a real scene is required. The results obtained will be used for teaching and research at the FEI STU in Bratislava.

Virtual laboratory based on Node.js technology

The paper demonstrates remote control of test experiment in the virtual laboratory. This is a common problem, but another way can always be used to solve it. The paper compares several existing virtual laboratories and their possible issues at present. To develop such a new solution JavaScript technology was used on both client and server side using Node.js runtime.

Proposal of system for automatic weld evaluation

The paper deals with the development of a system for automatic weld recognition using new information technologies based on cloud computing and single-board computer in the context of Industry 4.0. The proposed system is based on a visual system for weld recognition, and a neural network based on cloud computing for real-time weld evaluation, both implemented on a single-board low-cost computer. The proposed system was successfully verified on welding samples which correspond to a real welding process in the car production process. The system considerably contributes to the welds diagnostics in industrial processes of small- and medium-sized enterprises.

Application of Pattern Recognition for a Welding Process

The paper deals with the development of a system for automatic weld recognition using new information technologies based on cloud computing and single-board computer in the context of Industry 4.0. The proposed system is based on a visual system for weld recognition, and a neural network based on cloud computing for real-time weld evaluation, both implemented on a single-board low-cost computer. The proposed system was successfully verified on welding samples which correspond to a real welding process in the car production process. The system considerably contributes to the welds diagnostics in industrial processes of smalland medium-sized enterprises.

Weld segmentation for diagnostic and evaluation method

The work is devoted to the problem of weld segmentation in modern methods for evaluating the quality of welds. One of the methods of the weld quality evaluation is visual inspection. This work deals with the possibility of implementing algorithms of recognition and evaluation of welds using visual system in single-board computers. Result of this paper is computing time comparison of chosen embedded single-board computers and weld segmentation results based on entropy.