Istvan Szalay

56705163100

Publications - 9

Localization robustness improvement for an autonomous race car using multiple extended Kalman filters

Krisztian Enisz Ernö Horváth Istvan Szalay

Publication Name: Proceedings of the Institution of Mechanical Engineers Part D Journal of Automobile Engineering

Publication Date: 2025-08-01

Volume: 239

Issue: 9

Page Range: 3771-3783

Description:

In this paper, we introduce a vehicle localization method designed for the SZEnergy race car, which competes in the Shell Eco-marathon. The proposed method comprises four different extended Kalman filter-based localization algorithms and a selection algorithm that determines the most suitable one based on vehicle speed, GNSS availability, and signal quality. The low-speed Kalman filters are based on a kinematic vehicle model while the high-speed variants are based on a dynamic vehicle model. Several measurements were performed during test maneuvers to evaluate the performance of the filters. The proposed method succesfully handles sensor miscalibration and GNSS outages.

Open Access: Yes

DOI: 10.1177/09544070241266281

Simscape Implementation of a Nonlinear Permanent Magnet Synchronous Machine Model for Sensorless Polarity Detection

Istvan Szalay Denes Fodor

Publication Name: 2024 IEEE 21st International Power Electronics and Motion Control Conference Pemc 2024

Publication Date: 2024-01-01

Volume: Unknown

Issue: Unknown

Page Range: Unknown

Description:

This paper presents the implementation of a novel extended PMSM model in the Simscape acausal modeling subsystem of the widely used Mathworks MATLAB-Simulink numerical simulation environment. The extended PMSM model is based on a quadratic flux-current function that describes the machine's rotor position-dependent and polarity-dependent properties and facilitates the development of signal injection-based sensorless polarity detection algorithms. The implemented PMSM model was validated using simulations and measurements. The presented simulation results include modulated sinusoidal injection and non-modulated square-wave injection which are commonly used carrier signals in sensorless algorithms, and the simulation results were compared to corresponding measurements. A comparison of the acausal modeling approach in Simscape to the causal approach of Simulink is also presented. The implemented PMSM model is open source. The source code and the related data are available on GitHub.

Open Access: Yes

DOI: 10.1109/PEMC61721.2024.10726332

Embedded System Simulation Using Renode †

Istvan Szalay Ferenc Péter Speiser Denes Fodor

Publication Name: Engineering Proceedings

Publication Date: 2024-01-01

Volume: 79

Issue: 1

Page Range: Unknown

Description:

In the automotive industry, the reduction of development costs is of key importance. The development of electrical hardware is an expensive, time-consuming process with a lot of development stages (e.g., prototyping, electrical testing, mechanical testing, lifecycle testing). There is a growing need to increase the cost-effectiveness of the development and testing phases of embedded software using virtualization. Using this method, less prototype manufacturing is necessary since the simulations allow for faster and more effective discovery of a large portion of possible faults without building a hardware prototype. Renode is an open source embedded system simulation framework that facilitates software-based testing. The main goal of this paper is to explore the usability of the framework for automotive applications.

Open Access: Yes

DOI: 10.3390/engproc2024079052

Embedded System Simulation for Electrical Hardware Test Virtualization

Istvan Szalay Ferenc Péter Speiser Denes Fodor

Publication Name: Chemical Engineering Transactions

Publication Date: 2024-01-01

Volume: 114

Issue: Unknown

Page Range: 799-804

Description:

In the automotive industry, the reduction of development costs has a key importance. Also, the duration of the development is a significant aspect. Electrical hardware development is an expensive, time-consuming process with a lot of development stages (e.g., prototypes, electrical tests, mechanical tests, lifecycle tests). Virtualization is an important factor in making electrical component development more sustainable; using this method, less prototype manufacturing is necessary since the simulations make it faster and more effective to find out a large portion of possible faults without building a hardware prototype. The main goal of the paper is to explore the capabilities of seven software solutions that can manage the virtualization of the electrical development toolchain. The simulation environment should manage the simulation of the electrical circuit, including the sensors and loads, the microcontroller, and the execution and debugging of the uploaded program code. The study provides guidance on choosing the proper simulator depending on the simulation focus.

Open Access: Yes

DOI: 10.3303/CET24114134

Deep Learning-Based Approach for Autonomous Vehicle Localization: Application and Experimental Analysis

Krisztian Enisz Ernö Horváth Istvan Szalay Norbert Marko

Publication Name: Machines

Publication Date: 2023-12-01

Volume: 11

Issue: 12

Page Range: Unknown

Description:

In a vehicle, wheel speed sensors and inertial measurement units (IMUs) are present onboard, and their raw data can be used for localization estimation. Both wheel sensors and IMUs encounter challenges such as bias and measurement noise, which accumulate as errors over time. Even a slight inaccuracy or minor error can render the localization system unreliable and unusable in a matter of seconds. Traditional algorithms, such as the extended Kalman filter (EKF), have been applied for a long time in non-linear systems. These systems have white noise in both the system and in the estimation model. These approaches require deep knowledge of the non-linear noise characteristics of the sensors. On the other hand, as a subset of artificial intelligence (AI), neural network-based (NN) algorithms do not necessarily have these strict requirements. The current paper proposes an AI-based long short-term memory (LSTM) localization approach and evaluates its performance against the ground truth.

Open Access: Yes

DOI: 10.3390/machines11121079

Nonlinear PMSM Model Implementation in MATLAB-Simulink for Sensorless Polarity Detection

Istvan Szalay Denes Fodor

Publication Name: 2023 International Conference on Electrical Drives and Power Electronics Edpe 2023 Proceedings

Publication Date: 2023-01-01

Volume: Unknown

Issue: Unknown

Page Range: Unknown

Description:

This paper presents the implementation of a novel nonlinear PMSM model for signal injection-based sensorless polarity detection algorithm development. The model extension incorporates a quadratic flux-current function to describe the polarity-dependent saliency of the machine. The model was implemented in the Mathworks MATLAB-Simulink environment, which is widely used for numerical simulations of electric machines and drives. The presented simulation results include non-modulated square-wave injection and modulated sinusoidal injection which are commonly used in sensorless algorithms. Furthermore, these simulation results were compared to corresponding measurements, and the comparison between the simulated and measured data is also provided. The models are open source and available on GitHub.

Open Access: Yes

DOI: 10.1109/EDPE58625.2023.10273864

Convolutional Neural Network-Based Tire Pressure Monitoring System

Zoltan Marton Istvan Szalay Denes Fodor

Publication Name: IEEE Access

Publication Date: 2023-01-01

Volume: 11

Issue: Unknown

Page Range: 70317-70332

Description:

Tire pressure has a significant influence on the driving safety of road vehicles; therefore, it is mandatory in many countries to equip all new road vehicles with a tire pressure monitoring system (TPMS). There are two types of TPMSs in use: the direct TPMS (dTPMS) and the indirect TPMS (iTPMS), both of which have made significant improvement in the last decade. The most accurate iTPMS methods used in commercial vehicles apply the Fourier transform on wheel speed sensor (WSS) signals and extract the pressure-dependent eigenfrequency by utilizing center of gravity (CoG) or peak search (PS) methods, the research focus is shifting towards model-based and artificial intelligence-based methods. In this paper we propose a novel advanced iTPMS method based on modern signal processing and a convolutional neural network (CNN) for eigenfrequency detection. The proposed iTPMS method uses the hybrid wavelet-Fourier transform in combination with a CNN trained for pattern recognition-based eigenfrequency detection, and according to experimental results, it outperforms the commercially most frequently used Fourier transform and CoG method combination both in terms of computational requirement and accuracy.

Open Access: Yes

DOI: 10.1109/ACCESS.2023.3294408

Permanent Magnet Synchronous Motor Model Extension for High-Frequency Signal Injection-Based Sensorless Magnet Polarity Detection

Krisztian Enisz Istvan Szalay Hunor Medve Denes Fodor

Publication Name: Energies

Publication Date: 2022-02-01

Volume: 15

Issue: 3

Page Range: Unknown

Description:

In this paper, a novel extended permanent magnet synchronous motor model is presented that incorporates a quadratic flux-current function to represent the polarity-dependent saliency. The proposed model enables the design of sensorless polarity detection algorithms required by the initial position detection of permanent magnet synchronous motors. The novelty of the model is that it integrates the polarity-dependent saliency into the traditional machine model and introduces a new machine parameter, the polarity-dependent saliency coefficient, to specify the Hessian matrix of the flux-current function. A measurement method is presented for determination of the elements of the Hessian and the polarity-dependent saliency coefficient. The solution of the model is given for high-frequency sinusoidal pulsating voltage injection. Experimental results show that the proposed extended model accurately predicts the amplitudes and phases of the second harmonics of the motor currents, which are the carriers of the polarity-dependent information. This information enables a current measurement-based polarity detection algorithm using the phase difference between the fundamental and second harmonic of the apparent d-axis current. Both the presented measurement data and the proposed model show that injection in the d-direction is optimal for polarity detection.

Open Access: Yes

DOI: 10.3390/en15031131

Comparison of square-wave and sinusoidal signal injection in sensorless polarity detection for PMSMs

Krisztian Enisz Istvan Szalay Denes Fodor

Publication Name: 2022 IEEE 20th International Power Electronics and Motion Control Conference Pemc 2022

Publication Date: 2022-01-01

Volume: Unknown

Issue: Unknown

Page Range: 583-589

Description:

This paper compares the non-modulated square-wave voltage injection-based and sinusoidal voltage injection-based polarity detection methods applicable in sensorless PMSM drives. A quadratic PMSM model extension is presented that serves as a basis for the polarity detection techniques. The comparison includes the polarity-dependent current components produced by the injected signals, the design parameters related to the injected signal features and their effect on the probability of correct polarity detection. The performance differences and possible application areas are also discussed.

Open Access: Yes

DOI: 10.1109/PEMC51159.2022.9962876