The protection of critical infrastructure is increasingly challenged by evolving cyber threats and sophisticated attack methods. Traditional technical security systems, such as access control, alarm systems, and power supply monitors, often rely on outdated technologies and isolated data, limiting their ability to address modern risks or provide predictive insights. The SAFEGUARD project addresses these limitations by integrating Artificial Intelligence (AI) and Machine Learning (ML) into the digital transformation of security infrastructure. This paper presents a strategic framework for AI/ML adoption in critical infrastructure protection, including a 10-point plan for secure implementation. It explores the use of machine learning, natural language processing, and deep learning to enhance threat detection and system resilience, while also acknowledging associated risks and implementation challenges. A pilot study demonstrates how combining data from legacy systems (e.g., UPS, access control, CCTV, HVAC) with AI models can enable predictive analytics, improving detection of security incidents and operational anomalies. The results indicate enhanced risk awareness, faster incident response, and proactive maintenance capabilities. The paper concludes with key insights and outlines future directions for scaling the SAFEGUARD approach.

Magnetic levitation technology is widely used in various engineering applications, including transportation, industrial automation, and precision systems. A major challenge in controlling magnetic levitation systems is ensuring stability and precise positioning amidst nonlinear electromagnetic interactions and external disturbances. The PID controller plays a fundamental role in regulating the electromagnetic forces required for levitation. This study focuses on optimizing the PID controller for controlling a magnetic levitation system. The research examines mathematical modelling, simulation techniques, and experimental validation to enhance stability and precision. A comprehensive analysis of magnetostatic simulations and heuristic tuning methods is performed to improve control efficiency. Experimental results demonstrate that optimized PID tuning significantly improves the stability and responsiveness of the levitation system. The findings contribute to advancing control engineering by providing practical guidelines for refining PID control in dynamic and nonlinear environments.

The article aims to present a system capable of managing emergency power systems in an automated, digital, precise, economical and environmentally friendly manner, ensuring continuity of energy supply. Modern technological progress has resulted in an increased demand for advanced security systems capable of processing data in real-time. As devices become more and more integrated with electronic data collection and processing systems, the importance of energy autonomy in securing these systems in emergencies increases. This challenge is particularly important for entities managing extensive and geographically dispersed technical infrastructure, such as telecommunications, services railways, pipeline maintenance and military operations. Ensuring energy autonomy for such enterprises requires periodic technical assessment of energy storage solutions, especially batteries.

The article discusses the design process and research of RFI filters in Electromagnetic Compatibility (EMC) measurements concerning conducted disturbances for an induction heater. In the simulation studies, the Simulink package of Matlab software was used to create computer models of the filters. The experimental studies utilized an artificial network HM 6050-2 by Hameg and a spectrum analyzer HMS 3000, allowing for precise analysis of signals in the frequency domain ranging from 150kHz to 30MHz, which was essential for assessing device compliance with EMC standards. This study introduces an innovative approach to designing RFI filters for an induction heater, focusing on the effectiveness of suppressing conducted disturbances. The novelty of the research lies in the application of a combination of low-pass filters and common-mode chokes with different parameters, which allowed for compliance with stringent EMC standards. The work examined ten different RFI filters, three of which showed satisfactory filtering characteristics. The first-order filters under study did not meet the standards; according to simulations and EMC tests, it was found that as the inductance of the filter’s common-mode choke increased, its falling edge became steeper. Significantly better results were achieved with two- and three-stage filters, where the serial connection of simple filters resulted in a substantial reduction of conducted disturbances and, consequently, compliance with stringent standards. The design solutions along with electromagnetic compatibility studies were discussed in the paper.

This project aims to develop an innovative, open multi-domain early warning platform to monitor adverse events in water bodies and streams, such as salinity increase, acidification, temperature rise, chlorophyll content, etc. The platform involves monitoring dispersed measurement points, acquiring data from measurement stations in the form of autonomous buoys equipped with sensor sets tailored to individual needs. The platform enables remote and distributed monitoring of key locations in lakes and rivers for stakeholders, such as local governments, water managers, and businesses impacting water reservoirs. The open data aggregation system and the universal technological solution in the form of a measurement buoy will allow the acquisition of desired chemical and biological parameters, in line with the Water Framework Directive, using market-available and industry-recognized sensors. The planned buoy construction will accommodate both wired sensors directly connected to the measurement station and data acquisition from locally dispersed measurement points communicating wirelessly with the measurement station-buoy. With built-in machine learning and data analysis mechanisms, the platform will utilize a minimal number of measurement stations to achieve the desired level of data acquisition. The dispersion of sensors and the autonomy of measurement stations will ensure flexibility and scalability of measurements.

The article presents the control algorithm of the HRV unit which allows to obtain the balance of the supply and exhaust flows based on the tempera-tures measured at the heat exchanger. Based on experimental work, it has been proven that it is possible to balance air flows based on temperatures in any duct systems. It was also indicated that the ambient temperature of the HRV unit should be taken into account.

The paper investigates the design and performance of an active horizontal seat suspension driven by an induction motor, with the aim of reducing harmful vibrations and enhancing energy efficiency. The physical model employs a permanent magnet synchronous motor (PMSM) as a modern actuator to generate the active force. Both passive and active seat suspensions are examined, revealing that the active system significantly reduces resonant vibration by applying electromagnetic torque from an induction motor. The motoring/braking principle is explored, where braking an induction motor returns electric energy to the servo drive system, and regenerative braking is utilized to charge batteries. However, the current regenerative braking system recovers only 3% of power. A novel control design methodology is proposed for the subsequent project to enhance suspension dynamics, attenuate vibrations transmitted to the human body, and minimize energy consumption during operation. The proposed energy-regenerative electromagnetic actuator model, in conjunction with advanced programmable micro-controller algorithms and inverter switching schemes, aims to optimize energy recuperation and vibration energy harvesting in the vibration reduction system.

The article aims to examine the influence of various acoustic materials on the usable properties of headphones. Acoustic materials act as filters that modify the incident acoustic waves by attenuating specific frequency components of the acoustic signals. They can therefore be called acoustic filters because they affect the frequency response (bandwidth) of the headphones in question. Therefore, the choice of acoustic materials in headphones with a specific design can have a significant impact on their psychoacoustic properties. Calibrating your headphones is critical to ensure they don't amplify, attenuate or introduce any coloration to your frequencies. To address this problem, a dedicated test setup was designed and constructed to evaluate the frequency response of selected headphone models using seven different acoustic materials (acoustic filters). The final selection of the acoustic filter was made using an artificial neural network. The article presents the measurement methodology, and the obtained frequency response measurements were analyzed and compared.

The study aimed to design and then build a test station to measure the characteristics of headphones, where the effect of different acoustic materials used for filling the acoustic chamber was investigated. The first part of the paper provides a comparative review of existing commercial stations used to measure and determine the characteristics of headphones. The second part of the article presents the concept of the author's station, which was intended to be made using mainly a fast and efficient incremental method. Here, 3D printing was decided upon. Further on, the components of the built test station are described. A method of measurement was developed. Then the measured characteristics were presented, compared, and analyzed.

This paper presents an application for monitoring power systems in a Melex type vehicle. The application is intended to optimise the energy use of the modified hybrid drive system used on this vehicle. It was written in Python programming language and is used to display vehicle operating parameters on a touch screen located on the control panel. A system for monitoring voltages and currents, of the electrical part of the drive train of a Melex vehicle, was designed and built, together with a data acquisition system. Tests were carried out on the basis of the completed measuring system, a selected section of which is presented in this paper.

This paper proposes the author's design of a mobile robot intended for creating a two-dimensional map of the environment. The first part of the paper describes the scope of automated systems called robots. On this basis, the design assumptions of the author's robot were defined, the components used in the construction are shown, the operation of the Robot Operating System (ROS) and the packages used in this robot control system are presented. The main idea, which was to govern the project was its low manufacturing cost. Thus, the use of the optical system of the Kinect device. As a result. the operation of the algorithm controlling the robot via the ATmega328 microcontroller is demonstrated. The results of the experimental verification of the robot's performance are presented in the form of two-dimensional maps of the rooms in which the experiments were carried out.

In this paper a multi-degree of freedom model simulating biodynamic response of the seated human body is presented. The model consists of three masses connected by linear springs and dampers. The unknown model parameters are identified by means of the proposed optimisation procedure that minimises the error between numerically simulated and experimentally measured transmissibility functions over the chosen frequency range. Although the model involves significant simplifications due to the complexity of the human body system, very promising results are obtained for different directions of the vibration exposure, i.e. longitudinal x-direction, lateral y-direction and vertical z-direction. For this reason, the model can be easily implemented for modelling and simulating many problems related to the vibration safety of drivers and operators of working machines.

This paper presents issues related to selected human biomechanical models that have been and are being used to analyze human body motion, human interaction with the environment, vehicles, machines. The process of building a multi-segment human biomechanical model in Matlab-Simmechanics environment was presented. Anthropometric parameters of real human segments are introduced in the model. The selected exemplary movements that a human can perform, are presented using model. Results of simulations performed to obtain selected kinematic and dynamic data associated with these movements are shown. They indicate the tremendous analytical potential of these kind of models in various scientific aspects such as medicine, sports, and technology.

The paper presents the concept of the supporting structure and modeling of the main components of the rover, which can transport people and work in autonomous mode. This is ensured by an appropriate design of the rover's support frame and drive wheel system. The paper presents a computer model of selected fundamental drive components affecting the traction capabilities of the vehicle, namely the differential, frame, and wheels. The modeling was performed in a Matlab-SimMechanics environment. In this environment, the model was also assembled into a functional whole, which allows simulating the rover's operation in different terrain conditions. This is presented in the final part of the paper, where the simulation of overcoming an obstacle by the rover presented in this paper is presented. One of the studied cases, an urban obstacle in the form of stairs, was chosen. Displacement diagrams of individual structural elements of the rover prove its high traction and stability capabilities.

This paper presents the process of converting the bioelectrical sEMG signal into an analog signal, which is useful from the point of view of motor or actuator control. The constructed measurement path for signal acquisition enables examining the "raw" signal from sEMG sensors and tracing the entire path of signal processing up to its "useful" form, i.e., the analog signal reflecting the degree of contraction of a selected muscle. This paper presents the successive signal processing and filtering stages and discusses the electrical circuits used in the signal acquisition process.

The article deals with the problem of a closed space with vibro-acoustical source generating inside specific conditions, which form a field with some values p (acoustic pressure). In general, the field is described by specific time dependent variable w(t) in all points of space, whose location are defined by coordinate r(x,y,z). The second factor describing acoustic field is related to eigenfunctions Ψ(r), which are the solution of eigenvalues problem of considered space V. In this work, it is assumed that the field distribution can be described by using modal analysis assumptions. Therefore, the acoustic pressure is defined by a modal expansion i.e. the sum over a set of a space’s eigenfunctions Ψ(r) and time components w(t). The model indicates that inside the space there are source conditions (location and source character), which generate resonances strongly than others. In order to confirm this, the experiment was conducted and its results are shown. That was conducted using specific acoustic source placed inside chosen closed space. Next, using wavelet analysis, it is possible investigate the problem to specify areas, where the source generates unwanted resonances (correlated with object eigenfrequencies in particular frequency range) and their damping.

Experimental research set-up used to simulate the operation of the ABS system (Anti-Lock Braking System) is presented in the article. This experimental stand allows to show in an accessible way the principle of operation of the system preventing car traction loss. The paper presents measurements of the electrical signal transmitted from the inductive sensor to the ABS driver, depending on different vehicle speeds, different distance of the sensor from the plane of the rim’s tooth, and the most common damage to toothed rims. A method of identifying these damages has been proposed. This method is based on the wavelet analysis of signals.

The article presents a design solution of a mobile robot dedicated to autonomous movement in hard-to-reach places. The 3D printing technology made it possible to make a mechanical structure adapted to the actuators and electronic elements with which the robot was equipped. To control the mobile robot, a Raspberry Pi 2 microcomputer and an Atmega328P microcontroller were used, which allows the robot to be controlled via WiFi or a game controller for PlayStation3 using Bluetooth communication. In addition to remote control, the rover has an implemented algorithm of autonomous operation with simultaneous image transmission.