Marco Rossi (MathWorks)
Artificial intelligence (AI) is driving a massive change in the way engineers, scientists, and programmers develop and improve products and services. All engineering fields today use AI in one form or another, and many of today's industrial challenges call for engineers prepared to incorporate AI into their workflows. Find out how MathWorks tools empower engineers, including those with minimal AI experience, to develop better systems that use AI workflows. Additionally, we will discuss how to accelerate the incorporation of AI in engineering courses. You will walk away from this session with a better understanding of how a constant dialog between industry and academia can prepare engineers for the AI megatrend.
Petr Semotam (Siemens, s.r.o.)
The application processes industrial data such as signals and time series, text information from control systems or real-time operating time in combination with maintenance order records to provide information about the operability of the monitored machinery. Algorithms working with the data are built using applied statistics, neural networks or even the popular GPT model for working with text. The paper will give an insight into the background of web application development in practice and describe the techniques and procedures used to work with data.
Ivan Masár (TTTech Computertechnik)
TTEthernet is an Ethernet-based technology with a set of features providing Quality of Service (QoS) communications and security that was developed specifically for safety-critical systems. It is a network technology that extends standard Ethernet (IEEE802.3) with a protocol for real-time data transmission with defined latency. This makes it possible to integrate mission-critical data, for example for spacecraft control, with less critical data, such as video and audio, within a single data network.
Examples of applications of TTEthernet include the new European Ariane 6 launch vehicle, or the NASA/ESA manned lunar flight programmes Lunar Gateway and Artemis.
The design of avionics systems for such applications is a complex process. To simplify the design of real-time applications communicating over a TTEthernet network, we have designed and implemented development tools using MATLAB/Simulink to automatically generate the network configuration as well as the code for the applications themselves, which will be presented in the talk.
Martin Hrnčár (R-SYS s.r.o.)
The presentation deals with the problem of processing the output data from the Frequency Modulated Contious Wave (FMCW) radar network. In the application used, they serve as motion detection sensors for non-cooperative targets on airport taxiways. The traditional approach to FMCW radar data processing was not computationally efficient enough and the object detection results ran up against the limits of the algorithms used. Therefore, as part of a research and development project, we developed new algorithms using machine learning (ML) principles. The prototyping and verification of the algorithms was carried out in MATLAB environment. Verification and validation of the solution is carried out at the Bratislava airport in accordance with the requirements of the EUROCAE ED-116 standard. In the presentation, I will present the main challenges we had to face as well as the achievements we have made.
Ondrej Vacek (Technická univerzita vo Zvolene)
When processing wood as a highly heterogeneous inhomogeneous material interesting and non-trivial optimization problems arise. In this talk, we will mention some of the problems encountered when working with wood and introduce 3D CT scanning technology that allows non-invasive investigation of the properties of wood logs. We will show how we can use MatLab to evaluate scans and train neural networks to automatically detect defects in wood.
Radek Papoušek (Continental), Jakub Zábojník (Continental), Lubor Zháňal (Humusoft)
Humusoft offers consulting services and custom work related to the use of the tools we support. Services can be short-term consultations, but we also offer support for your product development, implementation of Model-Based Design, advanced business data processing and conversion or optimization of existing software.
As an example of a successful collaboration, we will present projects on which we have worked with Continental Barum.
These include software for advanced analysis of driving test data aimed at investigating tyre life through various wear and mileage measurements. And also auxiliary tools for managing and planning the use of internal resources.
Michal Blaho (Humusoft)
MATLAB helps researchers, developers and engineers create new algorithms and devices. It is published twice a year with many new features in the core module and individual extensions. During the presentation, you will see new capabilities for modeling, simulation, and design sharing, as well as new tools to increase your productivity and create better code and models. The new features will focus on areas such as:
Anna Tocháčková (Humusoft)
MATLAB offers a wide range of tools for visualizing different types of data. In addition to visualizing two-dimensional data, we can also visualize various forms of three-dimensional data in MATLAB:
Different charts are appropriate for each type of volume data. In this lecture we will give an overview of these graphs. Also related to the topic of visualizing three-dimensional data is the visualization of pointclouds (point clouds), which are used, for example, in robotics. MATLAB also supports working with image three-dimensional data such as segmentation, classification or 3D image reconstruction.
Jaroslav Jirkovský (Humusoft)
Use of the Model-Based Design method in the design, tuning and implementation of control systems. The method is based on dynamic system models created in Simulink, which are followed by the actual design of the control system. This can include simple control loops, such as PID control, as well as advanced types of control systems. Advanced options include robust tuning of control systems to include the effect of uncertain parameters, tuning of complex control systems with multiple inputs and outputs (MIMO), adaptive control systems (ADRC, ESC) or predictive control systems (MPC).
Jaroslav Jirkovský (Humusoft)
Model-Based Design development is based on the systematic use of simulation models throughout the development process. Typical steps in the development of an electric drive system model include modeling the electric motor (PMSM, BLDC, etc.), modeling the power electronics (voltage converter, inverter), designing and implementing embedded software (control algorithms, virtual sensors), defining test scenarios, simulating and analyzing the results. In terms of power source, it is possible to model battery cells, design different battery pack architectures and evaluate the thermal and electrical behaviour of batteries under normal and fault conditions. The aim of the battery management system (BMS) algorithms is then to ensure the required performance, safe operation and acceptable battery lifetime in different operating modes and under different ambient conditions. Finally, the development process relies on the automatic generation of source code (C/C++, HDL) from the proposed algorithms and its integration into the real environment.
Martin Kožíšek (Humusoft)
In this lecture, we will introduce the simulation tools COMSOL Multiphysics, COMSOL Server and COMSOL Compiler. We will look at the unique features of COMSOL software and show the ways in which modern precision simulation is advancing science and research in companies and scientific institutions.
Michal Blaho (Humusoft)
The creation of algorithms involves several steps from data retrieval, analysis and modification to the publication of the results. To facilitate these steps, MATLAB and the Live editor provide several applications and Live Tasks for convenient and fast work. These tools also provide the ability to generate code for easy step repetition and automation. During this talk, we will cover several tools useful for your work that you will discover in the latest versions of MATLAB. We will also introduce appropriate data types for data representation.
Tomáš Fridrich (Humusoft)
There are multiple ways to model and subsequently simulate power electronics on the dSPACE platform. Each method is independent and unique in its approach. Just as the methods of creating models are different, so too is the time required to create the models, as well as the level of detail in the resulting model. In this talk, you will learn when to use which method, what are the limits of the solutions, and what direction to take if you are just starting to develop engines or control algorithms.
PMSM, BLDC and other motors, as well as transmission modelling, can all be transferred to the dSPACE platform and modelled in real time.
Tomáš Vrbata (Humusoft)
You don't need a background in physics or mathematics for this workshop. Just install the COMSOL Multiphysics trial license in advance and you can set up the fluid flow simulation step-by-step together with our colleagues. Don't worry if something goes wrong, just raise your hand and we will help you.