Alexandros Soumelidis

6603122285

Publications - 8

Brake Disc Deformation Detection Using Intuitive Feature Extraction and Machine Learning

István Lakatos Tamás Dózsa Erno Simonyi Péter Ori Mátyás Szabari Alexandros Soumelidis

Publication Name: Machines

Publication Date: 2024-04-01

Volume: 12

Issue: 4

Page Range: Unknown

Description:

In this work we propose proof-of-concept methods to detect malfunctions of the braking system in passenger vehicles. In particular, we investigate the problem of detecting deformations of the brake disc based on data recorded by acceleration sensors mounted on the suspension of the vehicle. Our core hypothesis is that these signals contain vibrations caused by brake disc deformation. Since faults of this kind are typically monitored by the driver of the vehicle, the development of automatic fault-detection systems becomes more important with the rise of autonomous driving. In addition, the new brake boosters separate the brake pedal from the hydraulic system which results in less significant effects on the brake pedal force. Our paper offers two important contributions. Firstly, we provide a detailed description of our novel measurement scheme, the type and placement of the used sensors, signal acquisition and data characteristics. Then, in the second part of our paper we detail mathematically justified signal representations and different algorithms to distinguish between deformed and normal brake discs. For the proper understanding of the phenomenon, different brake discs were used with measured runout values. Since, in addition to brake disc deformation, the vibrations recorded by our accelerometers are nonlinearly dependent on a number of factors (such as the velocity, suspension, tire pressure, etc.), data-driven models are considered. Through experiments, we show that the proposed methods can be used to recognize faults in the braking system caused by brake disc deformation.

Open Access: Yes

DOI: 10.3390/machines12040214

Identification of the nonlinear steering dynamics of an autonomous vehicle

Gabor Rodonyi R. Tóth M. Schoukens Peter Koros Dániel Pup Ádám Kisari G. I. Beintema Zsombor Vigh Alexandros Soumelidis J. Bokor

Publication Name: IFAC Papersonline

Publication Date: 2021-07-01

Volume: 54

Issue: 7

Page Range: 708-713

Description:

Automated driving applications require accurate vehicle specific models to precisely predict and control the motion dynamics. However, modern vehicles have a wide array of digital and mechatronic components that are difficult to model, manufactures do not disclose all details required for modelling and even existing models of subcomponents require coefficient estimation to match the specific characteristics of each vehicle and their change over time. Hence, it is attractive to use data-driven modelling to capture the relevant vehicle dynamics and synthesise model-based control solutions. In this paper, we address identification of the steering system of an autonomous car based on measured data. We show that the underlying dynamics are highly nonlinear and challenging to be captured, necessitating the use of data-driven methods that fuse the approximation capabilities of learning and the efficiency of dynamic system identification. We demonstrate that such a neural network based subspace-encoder method can successfully capture the underlying dynamics while other methods fall short to provide reliable results.

Open Access: Yes

DOI: 10.1016/j.ifacol.2021.08.444

Characterization of Model Uncertainty Features Relevant to Model Predictive Control of Lateral Vehicle Dynamics

Gabor Rodonyi Dániel Pup Ádám Kisari Zsombor Vigh Alexandros Soumelidis J. Bokor

Publication Name: 2020 23rd IEEE International Symposium on Measurement and Control in Robotics Ismcr 2020

Publication Date: 2020-10-15

Volume: Unknown

Issue: Unknown

Page Range: Unknown

Description:

The information about a system's dynamics represented by measurement data sets are often confined to regions of restricted operations where the system is not sufficiently excited for model identification purposes. Experiments performed in closed-loop with safety constraints allow only for reduced order modeling. In the paper, a set of low order models are identified from real experimental data of the lateral dynamics of an electric passenger car. Low order models are advantageous for on-line computation in model-based control, though uncertainty due to neglected dynamics may deteriorate control performance and constraint satisfaction. The effect of uncertainty is analyzed by controller cross-validation where a controller designed based on one model is evaluated on other models playing the role of the true system. This method allows us to qualify not only model-controller pairs, but to determine the properties of input data and model uncertainty, which lead to more useful data sets, more robust and better performing controllers than the others.

Open Access: Yes

DOI: 10.1109/ISMCR51255.2020.9263745

Comparison of High-Precision GNSS systems for development of an autonomous localization system

Huba Somogyi Alexandros Soumelidis

Publication Name: 2020 23rd IEEE International Symposium on Measurement and Control in Robotics Ismcr 2020

Publication Date: 2020-10-15

Volume: Unknown

Issue: Unknown

Page Range: Unknown

Description:

In the near future, the vehicles offering advanced driver assistance or fully autonomous operation. It will require increasingly accurate position information, available in all driving conditions and with 100 percent availability. In generally, a single sensor cannot meet these requirements alone and so it is necessary to use a combined sensor suite solution incorporating different kinds of sensors working together. As the only source of absolute position, velocity and time, Global Navigation Satellite systems (GNSS) play a critical role in next generation positioning systems. In this paper, we are dealing with modern GNSS solutions and presents the results of the comparison of two high-precision GNSS systems in different driving conditions and presents conclusion for the develop of autonomous localization system.

Open Access: Yes

DOI: 10.1109/ISMCR51255.2020.9263762

Experimental verification of a control system for autonomous navigation

Gabor Rodonyi Peter Koros Dániel Pup Ádám Kisari Benedek Szucs Alexandros Soumelidis J. Bokor

Publication Name: IFAC Papersonline

Publication Date: 2020-01-01

Volume: 53

Issue: 2

Page Range: 14273-14278

Description:

A flexible architecture is developed with the purpose of supporting education and research on the field of autonomous vehicles. A pure electric vehicle is equipped with on-board computational units, sensors and actuator interfaces. This paper presents the current status of the control system and its validation by means of navigation experiments. With the cascade control architecture, problems of actuator dead-zone, sensor offset errors, path tracking and redesign for obstacle avoidance are addressed.

Open Access: Yes

DOI: 10.1016/j.ifacol.2020.12.1171

Research of required vehicle system parameters and sensor systems for autonomous vehicle control

András Mihály Peter Koros Dániel Pup Huba Somogyi Alexandros Soumelidis

Publication Name: Saci 2018 IEEE 12th International Symposium on Applied Computational Intelligence and Informatics Proceedings

Publication Date: 2018-08-20

Volume: Unknown

Issue: Unknown

Page Range: 27-32

Description:

Our long-term goal is to implement autonomous vehicle control functions on a standard vehicle. At first we started with the investigation of the steering, which is a crucial area of the control of autonomous vehicles. As a part of the program, vehicle dynamical measurements were carried out on a Nissan Leaf electric vehicle equipped with a sensor system, furthermore we design a high-level trajectory-tracking controller.

Open Access: Yes

DOI: 10.1109/SACI.2018.8441008

Structure Selection and LPV Model Identification of a Car Steering Dynamics⁎

Gabor Rodonyi Zoltán Szabó Peter Koros Dániel Pup Alexandros Soumelidis J. Bokor

Publication Name: Unknown

Publication Date: 2018-01-01

Volume: 51

Issue: 15

Page Range: 1086-1091

Description:

A Linear Parameter-Varying (LPV), discrete-time black box model of an electric power assisted steering system of a passenger car is identified from open-loop step response measurement data. The goal is to provide a nominal model for control design and analysis that is able to describe the principal characteristics of the system in the whole region of steering angle and speed range of 3 to 30 km/h. Examining a set of experimental data by using classical linear time-invariant black box modeling and validation techniques, the structure of the LPV model is determined. The parameters of the model are identified based on minimizing a quadratic error criterion by nonlinear optimization algorithms.

Open Access: Yes

DOI: 10.1016/j.ifacol.2018.09.049

System identification with generalized Prony schemes

Ferenc Schipp Tamás Dózsa Bokor Jozsef Gergo Ungvari Alexandros Soumelidis

Publication Name: Proceedings of the American Control Conference

Publication Date: 2025-01-01

Volume: Unknown

Issue: Unknown

Page Range: 5086-5092

Description:

We propose a novel method to identify the transfer functions of single-input-single-output linear time invariant (SISO-LTI) dynamic systems. Our approach makes use of the operator based generalization of Prony's method. In particular, the operator based Prony algorithm is used to reconstruct the transfer function of the system as a linear combination of rational basis functions. A considerable benefit of the proposed method is its robustness against the estimated system order. That is, if system order is over estimated, the correct system order can be found naturally. Another important benefit is that the proposed method is shown to be asymptotically robust towards zero expectation noise with the correct choice of certain evaluation functionals. The effectiveness of the proposed method is demonstrated through numerical experiments.

Open Access: Yes

DOI: 10.23919/ACC63710.2025.11107575