Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Piotr Przystałka has a total of 38 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Quantitative Evaluation of Thermomechanical Fatigue Damage in Composite Materials Using X-Ray Computed Tomography and Machine Learning

    by Piotr Przystałka, Paweł Chrzanowski, Tomasz Rogala, Andrzej Katunin

    The fatigue degradation of polymer-matrix composite materials, especially in the case of thermomechanical fatigue, when the self-heating effect dominates the process of the cyclically bent structure in fully reversed mode, results in complex damage patterns with a heterogeneous nature. It is therefore essential to evaluate the structural condition of tested objects in a non-destructive way to predict their fatigue lifetime. One of the most accurate non-destructive testing techniques used for such examinations is X-ray computed tomography, which enables the evaluation of the resulting fatigue damage in a three-dimensional space. However, the main challenge in the evaluation of such results is the attribution of the observed damage to specific types, the assessment of its severity, and its quantitative analysis. In this study, we have proposed a novel framework based on intensity and morphological features for the classification of damage types and a 3D U-Net-based approach for segmentation. The acquired results demonstrated the high accuracy of the performance evaluation indices, which exceeded 99% for the proposed framework, allowing effective evaluation of the resulting damage and being potentially helpful in the assessment of structural performance and residual fatigue life. Such analysis contributes to the trends of NDE 4.0 and adaptive mechanics, where composite structures are tailored to specific operational conditions to reach the target lifecycle.

  • Open Access

    Physics-Informed Neural Networks in Aerospace Engineering: A Systematic Review of Architectures, Training Strategies, and Open Challenges

    by Przemysław Gryt, Piotr Przystałka

    This paper provides a systematic synthesis of recent developments in physics-informed neural networks (PINNs) applied to aerospace engineering, with an emphasis on their role in physically consistent surrogate modeling, forward simulation, and inverse parameter estimation. Using a PRISMA-based methodology, the study surveys peer-reviewed works published between 2017 and 2025 across aviation- and space-related domains, including aerodynamics, structural mechanics, aeroelasticity, propulsion, control, structural health monitoring, satellite-orbit prediction, space-debris collision avoidance, and spacecraft radiation-impact modeling. The analysis shows that embedding governing equations, boundary conditions, and observational data into composite loss functions enables PINNs to improve predictive consistency, reduce dependence on dense simulation or experimental datasets, and support parameter identification under sparse or noisy measurements. Attention is given to architectural variants such as XPINNs, cPINNs, gPINNs, operator-learning approaches, and hybrid PINN-CFD/FEM formulations, as well as to training strategies based on adaptive sampling, domain decomposition, transfer learning, and dynamic loss weighting. Reported benefits include reduced approximation error, improved convergence in selected high-gradient or multiphysics problems, and enhanced interpretability compared with purely data-driven models. At the same time, the review identifies persistent open challenges, including scalability to large aerospace domains, sensitivity to loss-weighting and collocation strategies, limited robustness under noise and uncertainty, high computational cost, and the lack of standardized aerospace benchmarks. Overall, the review highlights PINNs as a promising but still developing framework for fast, interpretable, and physically consistent modeling of aircraft and spacecraft systems.


  • Open Access

    Regression-Based Modeling for Energy Demand Prediction in a Prototype Retail Manipulator

    by Piotr Kroczek, Krzysztof Lis, Piotr Przystałka

    The present study proposes two regression-based models for predicting the energy consumption of a four-axis prototype retail manipulator. These models are developed using experimental current and voltage measurements. The Total Energy Model (TEM) is a method of estimating energy per trajectory that utilizes global motion parameters. In contrast, the Power-to-Energy Model (PEM) is a technique that reconstructs energy from predicted instantaneous power. It has been demonstrated that both models demonstrate high levels of predictive accuracy, with mean absolute percentage error (MAPE) values ranging from 1 to 1.5%. These models are well-suited for implementation in hardware-constrained environments and for integration into digital twins.


  • Open Access

    Towards Safer Electric Vehicles: Autoencoder-Based Fault Detection Method for High-Voltage Lithium-Ion Battery Packs

    by Grzegorz Wójcik, Piotr Przystałka

    The rapid growth in the battery electric vehicle (BEV) market has brought lithium-ion battery (LIB) packs to the forefront due to their superior power and energy density properties. However, LIBs are highly susceptible to environmental factors, operating conditions, and manufacturing inconsistencies and operate within a narrow safety operating window. Battery faults pose significant risks, including potentially catastrophic thermal runaway, that can be initiated even by small faults, propagating further into a chain reaction cascade of failures. Aiming to improve the safety of such battery packs, this article presents the developed autoencoder-based fault detection method. The method, enhanced by computational intelligence and machine learning, is a result of extensive research into optical liquid detection systems (OLDSs) for immersion-cooled battery packs, where optical rather than electrical signals are used inside high-voltage areas. The performance was evaluated using recorded real-life datasets under faultless states and under simulated fault states through specific model performance indicators as well as detection performance indicators.


  • Open Access

    Adaptive Path Planning for UAV-Based Pollution Sampling

    by Mateusz Kosior, Piotr Przystałka, Wawrzyniec Panfil

    Unmanned Aerial Vehicles (UAVs) continue to gain popularity in applications such as military reconnaissance, environmental monitoring in remote locations, and package delivery. High-Altitude Long-Endurance (HALE) UAVs can remain airborne for extended periods, enabling air pollution measurements to be conducted across a wide range of altitudes, from a few hundred meters above ground level to the lower stratosphere. However, the challenges posed by dynamic environmental conditions and strict energy limitations necessitate the use of adaptive path planning algorithms that account for UAV and environmental models. To address these challenges, we propose a two-tier Adaptive Path Planner (APP), which comprises a Global Path Planner (GPP) and a Local Path Planner (LPP). The GPP, operating offline, generates obstacle-free, energy-efficient paths that adhere to the UAV’s kinematic constraints, while the LPP dynamically recalculates alternative routes in real time when obstacles arise. The APP leverages a novel data-driven environmental model, integrating terrain, wind, airspace, and measurement maps. Extensive Model-in-the-Loop testing was conducted to evaluate various single-objective optimization algorithms for the GPP. Subsequently, the APP was successfully validated in simulation scenarios inspired by real-world pollution monitoring missions conducted in Poland and the Arctic. Additionally, the proposed approach was tested under real-world conditions, demonstrating significant application potential. A comparative analysis of the generated paths demonstrated that the APP effectively replaces human operators. Further testing confirmed the APP’s capability for adaptive re-planning during mission execution.


  • Open Access

    Automating the Analysis of Negative Test Verdicts: A Future-Forward Approach Supported by Augmented Intelligence Algorithms

    by Anna Gnacy-Gajdzik, Piotr Przystałka

    In the epoch characterized by the anticipation of autonomous vehicles, the quality of the embedded system software, its reliability, safety, and security is significant. The testing of embedded software is an increasingly significant element of the development process. The application of artificial intelligence (AI) algorithms in the process of testing embedded software in vehicles constitutes a significant area of both research and practical consideration, arising from the escalating complexity of these systems. This paper presents the preliminary development of the AVESYS framework which facilitates the application of open-source artificial intelligence algorithms in the embedded system testing process. The aim of this work is to evaluate its effectiveness in identifying anomalies in the test environment that could potentially affect testing results. The raw data from the test environment, mainly communication signals and readings from temperature, as well as current and voltage sensors are pre-processed and used to train machine learning models. A verification study is carried out, proving the high practical potential of the application of AI algorithms in embedded software testing.


  • Detection of Cyber Attacks in Electric Vehicles Using a Deep Neural Network

    by Marcin Gajdzik, Anna Timofiejczuk, Anna Gnacy-Gajdzik, Piotr Przystałka

  • A Model-Based Approach for Testing Automotive Embedded Systems – A Preliminary Study

    by Anna Gnacy–Gajdzik, Marcin Gajdzik, Piotr Przystałka, Kamil Sternal

  • Open Access

    Development of a Power and Communication Bus Using HIL and Computational Intelligence

    by Marek Sznura, Piotr Przystałka

    This paper deals with the development of a power and communication bus named DLN (Device Lightweight Network) that can be seen as a new interface with auto-addressing functionality to transfer power and data by means of two wires in modern cars. The main research goal of this paper is to elaborate a new method based on a hardware in the loop technique aided by computational intelligence algorithms in order to search for the optimal structure of the communication modules, as well as optimal features of hardware parts and the values of software parameters. The desired properties of communication modules, which have a strong influence on the performance of the bus, cannot be found using a classical engineering approach due to the large number of possible combinations of configuration of the hardware and software parts of the whole system. Therefore, an HIL-based optimization method for bus prototyping is proposed, in which the optimization task is formulated as a multi-criteria optimization problem. Several criterion functions are proposed, corresponding to the automotive objectives and requirements. Different soft computing optimization algorithms, such as a single-objective/multi-objectives evolutionary algorithm and a particle swarm optimization algorithm, are applied to searching for the optimal solution. The verification study was carried out in order to show the merits and limitations of the proposed approach. Attention was also paid to the problem of the selection of the behavioural parameters of the heuristic algorithms. The overall results proved the high practical potential of the DLN, which was developed using the proposed optimization method.