Ján Liguš

Cybernetic education centre

Cybernetic education centre (CEDUC) integrates (1) the e-learning system Moodle, which primary provides the linear delivery of knowledge content managed by teacher as well as collaborative e-learning modules, (2) distributed real, virtual, simulation models as a support for e-learning to give student interactive feedback within the learning process and (3) the Assessment module realized in the in the Learning Management System (LMS) Moodle. The e-learning in Department of Cybernetics and Artificial Intelligence is not full online learning system, but all subjects are implemented as blended learning combining classical face-to-face learning with e-learning. This paper deals with remote access to controlled models in the laboratories of our department, (a) real laboratory models, (b) simulation models and (c) virtual models, whose implement many technologies from world leaders in area of automation and can not be implemented natively as Moodle modules. Remote student access to control process of these models is conditional, depending on their assessment results. The paper presents methodology of design of blended learning using remote control of models, in are of Cybernetic and Automation, implemented in CEDUC.

Cybernetic Education Centre: Monitoring and control of learner's e-learning study in the field of Cybernetics and Automation by Coloured Petri Nets model

This paper presents the Cybernetic Education Centre (CEDUC) as a hybrid e-learning and training centre for higher education of Cybernetics and Automation fields. If we consider Cybernetics we consider basically (1) controlled systems and (2) control systems. In case of controlled systems learner is focused on the process of analyze, identification, design of the mathematical model and simulation of the controlled system. Therefore this paper deals with controlled models in the laboratories of our department (a) real laboratory models, (b) simulation models and (c) virtual models which creates one integrated hybrid architecture what represents one of new ideas in the paper (Fig. 1). Learner of control systems is focused mainly on design of control parameters, design of control algorithms, design of hardware, software and communication architectures of control systems. Overall control system is represented by Distributed Control System (DCS) which serves for learners to verify designed control systems. The verification of the control systems is very important from safety point of view to prepare learners for real production conditions. Second new idea of the paper is implementation of the Coloured Petri Nets as automata to control access to the resources (not only typical study resources but also access to the components of hybrid DCS architecture) as well as to monitor the learner activities during the study. CEDUC is supported by intensive industry-university partnership. Conclusion of the paper summarizes the results of the study process of learners in DCS environment.

Reliability increasing through networked cascade control structure – consideration of quasi-redundant subsystems

Abstract The paper presents common cascade control architecture where specific kind of redundancy could be considered. There are different approaches how to increase the reliability of networked control systems. Common approach uses redundant components in control system i.e. passive or active redundancy. We deal with quasi-redundant subsystems (shared redundancy) which could be very useful when critical failure appears. This type of redundancy offers several important advantages such as minimizing the number of components as well as reliability increasing. The example of a four-rotor mini-helicopter is presented where a cascade control architecture is considered. The main contribution of this paper is that it helps to determine the applications where quasi-redundant subsystems are a good solution to remain in a significant reliability level even if critical failure appears.

Systematic design of the web portals

If we want simulation full-dynamic system than we must use evolution process let us say life cycle. The system during life cycle is passed by from simple deterministic system to large scale stochastic system. Life cycle of system consist of several phases with contains inside process such as remove relation between component, adding relation between component, adding and removing component and so on and so on.

Cooperation between industry and university based on the evaluation of the industrial research results in the academic environment

Based on the European Structural Funds it was developed the Intelligent Mobile Box, Intelligent Panel Controller with intelligent adaptive controllers within the industrial research and experimental development in the company Kybernetes, s.r.o. Within the frame of the academic-industry cooperation, the intelligent adaptive controller was tested at the Department of Cybernetics and Artificial Intelligence, Technical university of Kosice, Slovakia. The tests of the mobile intelligent adaptive controller were performed on two levels of university study, on the Bachelor level on the exercises from the subject “Control of Technological Processes” and on the Engineering level the exercises from the subject “Intelligent Control Networks” and on one Diploma project. Goals of students of the Control of the technological processes course had two goals, firstly to connect the intelligent adaptive controller to pre-defined controlled system (real plant, real model or simulated model) and next to validate the control results. Students of the Diploma project on the Engineering level had more advanced goals. Tasks defined for engineering students were to connect the intelligent adaptive controller to non-defined controlled system, setup the adaptivity process of the controller regarding the learning error, parameterize the control system, observe and validate the control results. Both sides concluded this cooperation as very valuable. Main contributions for students were (U1) the challenge to apply studied theoretical knowledge on the real industrial controllers, (U2) experience with new research results and technologies deployed in industry and (U3) the implementation of the control and adaptive algorithms from abstract mathematical area to real PLC controller. On side of industry research company the main contributions were (C1) testing of designed algorithms and (C2) user feedback from students to make the application HMI interface more understandable a native.

Study of different load dependencies among shared redundant systems

The paper presents features and implementation of a shared redundant approach to increase the reliability of networked control systems. Common approaches based on redundant components in control system use passive or active redundancy. We deal with quasi-redundant subsystems (shared redundancy) whereas basic features are introduced in the paper. This type of redundancy offers several important advantages such as minimizing the number of components as well as increasing the reliability. The example of a four-rotor mini-helicopter is presented in order to show reliability improving without using any additional redundant components. The main aim of this paper is to show the influence of the load increasing following different scenarios. The results could help to determine the applications where quasi-redundant subsystems are a good solution to remain in a significant reliability level even if critical failure appears.

The attributed importance in the humanitarian dimension of study of technological direction from the factors of technological education

In this work is studied the attributed importance in the humanitarian dimension of study in Electrical and Information Engineering, the prominent technological studies, from the factors and others involved in the education in Europe (academic, employer, graduated, student). The conclusions are based on the statistical assessment of roughly 2000 answers on relative questions, inside the frame of a more general research, for the assessment of the quality of programs of study on the corresponding speciality, which took place during the Thematic Network ldquoSurveyorrdquo.

Modification of the Smith Predictor for Random Delays Treatment in the Network Control Systems

Abstract This paper discusses analysis and design of control algorithms in Network Control Systems (NCS). The problem analysed is the treatment of systems with random time delays caused by a communication network. The main idea of this paper is to consider network induced delay as a dynamically changeable deadtime. In regard to this assumption, methods and algorithms known from area of control systems with deadtime can be used to compensate influence of network delays in NCS. In our case the method of the Smith predictor was modelled in actuator node. The modelling of NCS is realized by Coloured Petri nets (CPN) and as simulation tool is used Design/CPN software.

Decentralized Decision and Control

Abstract In this paper the methods of decentralized quantitative (statistical) and qualitative (fuzzy) decision and control algorithms are presented. The body of the paper is subdivided into two main sections. The first one describes decentralized statistical decision algorithms based on Bayesian conditional probabilities and the second one describes decentralized fuzzy control and decision algorithms. The derived mathematical apparatusies are described in detail. The principles of decentralization methods are designed and verified by simulations. The pendulum-car system and dynamic systems with transfer functions:are used as application models.