TY - GEN AU - Tettamanti, Tamás AU - Chinedu, Amabel Okolie AU - Ormándi, Tamás AU - Varga, Balázs TI - A Python tool for SUMO traffic simulation to model Hybrid Electric Vehicles (HEV) for comprehensive emission analysis PY - 2024 SP - 1 EP - 22 PG - 22 DO - 10.13140/RG.2.2.33734.69443 UR - https://m2.mtmt.hu/api/publication/34827436 ID - 34827436 LA - English DB - MTMT ER - TY - JOUR AU - Szőke, László AU - Shperberg, Shahaf S. AU - Holtz, Jarrett AU - Allievi, Alessandro TI - Adaptive Curriculum Learning with Successor Features for Imbalanced Compositional Reward Functions JF - IEEE ROBOTICS AND AUTOMATION LETTERS J2 - IEEE ROBOT AUTOM LETT PY - 2024 PG - 8 SN - 2377-3766 DO - 10.1109/LRA.2024.3387134 UR - https://m2.mtmt.hu/api/publication/34803158 ID - 34803158 N1 - Budapest University of Technology and Economics, Hungary Ben-Gurion University, Israel Robert Bosch LLC., USA Export Date: 22 April 2024 AB - This work addresses the challenge of reinforcement learning with reward functions that feature highly imbalanced components in terms of importance and scale. Reinforcement learning algorithms generally struggle to handle such imbalanced reward functions effectively. Consequently, they often converge to suboptimal policies that favor only the dominant reward component. For example, agents might adopt passive strategies, avoiding any action to evade potentially unsafe outcomes entirely. To mitigate the adverse effects of imbalanced reward functions, we introduce a curriculum learning approach based on the successor features representation. This novel approach enables our learning system to acquire policies that take into account all reward components, allowing for a more balanced and versatile decision-making process. IEEE LA - English DB - MTMT ER - TY - GEN AU - Tettamanti, Tamás AU - Ormándi, Tamás AU - Wágner, Tamás AU - Varga, Balázs AU - Varga, István TI - Korszerű eszközök a közúti járműforgalom modellezésére és irányítására. Bemutatkozik a BME Traffic Lab TS - Bemutatkozik a BME Traffic Lab PY - 2024 SP - 34 EP - 37 PG - 4 UR - https://m2.mtmt.hu/api/publication/34791003 ID - 34791003 AB - A 2007 óta működő BME Traffic Lab kutatócsoport kutat, fejleszt és nem utolsósorban oktatási tevékenységet is ellát a közúti járműforgalom mérése, modellezése, valamint irányítása területén. Munkája során az alapvető matematikai forgalommodellezésből kiindulva, illetve korszerű forgalomszimulációs eljárások alkalmazásával vizsgálja és fejleszti a különböző forgalomirányító algoritmusokat, egészen a forgalomirányító rendszerek programozásáig. Az elmúlt 10 évben a kutatásai kiegészültek az autonóm járművek forgalomra gyakorolt hatásainak vizsgálatával is, valamint természetszerűen a mesterségesintelligencia-alkalmazások is beszűrődtek a feladatokba. A BME Traffic Lab folyamatosan részt vesz hazai és nemzetközi K+F-projektekben, frissen tartva a szaktudást, és garantálva, hogy az elmélet a gyakorlattal is találkozzon. LA - Hungarian DB - MTMT ER - TY - JOUR AU - Tesone, Alessio AU - Tettamanti, Tamás AU - Varga, Balázs AU - Bifulco, Gennaro Nicola AU - Pariota, Luigi TI - Multiobjective Model Predictive Control Based on Urban and Emission Macroscopic Fundamental Diagrams JF - IEEE ACCESS J2 - IEEE ACCESS VL - 12 PY - 2024 SP - 52583 EP - 52602 PG - 20 SN - 2169-3536 DO - 10.1109/ACCESS.2024.3387664 UR - https://m2.mtmt.hu/api/publication/34788066 ID - 34788066 AB - Increasing motorization represents a severe problem worldwide, also affecting the emission levels of the road network. Accordingly, congestion management has obtained growing importance because of its strong economic, social, and environmental implications. Macroscopic Fundamental Diagram (MFD) based traffic control is a popular and efficient approach in this scientific field. In our research work the urban network has been divided into homogeneous regions, each of them characterized by its own MFD, and they are regulated using a network-level control scheme. The proposed Multiobjective Model Predictive Control (M-MPC) takes into account the congestion and CO2 emission levels of the urban network, modelled by the emerging Emission Macroscopic Fundamental Diagram (e-MFD). The applied strategy has been demonstrated in a realistic traffic scenario (Luxembourg City) using validated microscopic traffic simulation. According to the introduced multiobjective approach, the control method can better exploit the road network capacity while efficiently reducing traffic-induced emissions. Authors LA - English DB - MTMT ER - TY - JOUR AU - Szabó, Ádám AU - Doba, Dániel Károly AU - Aradi, Szilárd AU - Kiss, Péter TI - Model Development for Off-Road Traction Control: A Linear Parameter-Varying Approach JF - AGRICULTURE-BASEL J2 - AGRICULTURE-BASEL VL - 14 PY - 2024 IS - 3 PG - 16 SN - 2077-0472 DO - 10.3390/agriculture14030499 UR - https://m2.mtmt.hu/api/publication/34756334 ID - 34756334 N1 - Department of Control for Transportation and Vehicle Systems, Faculty of Transportation Engineering and Vehicle Engineering, Budapest University of Technology and Economics, Műegyetem rkp. 3., Budapest, H-1111, Hungary Institute of Technology, Department of Vehicle Technology, Hungarian University of Agriculture and Life Sciences, Páter Károly u. 1., Gödöllő, H-2100, Hungary Export Date: 5 April 2024 Correspondence Address: Aradi, S.; Department of Control for Transportation and Vehicle Systems, Műegyetem rkp. 3., Hungary; email: aradi.szilard@kjk.bme.hu AB - The number of highly automated machines in the agricultural sector has increased rapidly in recent years. To reduce their fuel consumption, and thus their emission and operational cost, the performance of such machines must be optimized. The running gear–terrain interaction heavily affects the behavior of the vehicle; therefore, off-road traction control algorithms must effectively handle this nonlinear phenomenon. This paper proposes a linear parameter-varying model that retains the generality of semiempirical models while supporting the development of real-time state observers and control algorithms. First, the model is derived from the Bekker–Wong model for the theoretical case of a single wheel; then, it is generalized to describe the behavior of vehicles with an arbitrary number of wheels. The proposed model is validated using an open-source multiphysics simulation engine and experimental measurements. According to the validated results, it performs satisfactorily overall in terms of model complexity, calculation cost, and accuracy, confirming its applicability. LA - English DB - MTMT ER - TY - JOUR AU - Imran, Waheed AU - Tettamanti, Tamás AU - Varga, Balázs AU - Bifulco, Gennaro Nicola AU - Pariota, Luigi TI - Macroscopic modeling of connected, autonomous and human-driven vehicles: A pragmatic perspective JF - TRANSPORTATION RESEARCH INTERDISCIPLINARY PERSPECTIVES J2 - TRANSPORT RES INTERDISC PERSP VL - 24 PY - 2024 PG - 17 SN - 2590-1982 DO - 10.1016/j.trip.2024.101058 UR - https://m2.mtmt.hu/api/publication/34724815 ID - 34724815 N1 - Department of Civil, Architectural and Environmental Engineering, University of Naples, Federico II, Via Claudio 21, Campania, Naples, 80125, Italy Department of Control for Transportation and Vehicle Systems, Faculty of Transportation Engineering and Vehicle Engineering, Budapest University of Technology and Economics, Műegyetem rkp. 3., Budapest, H-1111, Hungary Export Date: 18 March 2024 Correspondence Address: Pariota, L.; Department of Civil, Federico II, Via Claudio 21, Campania, Italy; email: luigi.pariota@unina.it Funding details: 2020Z9HEMJ Funding details: RRF-2.3.1-21-2022-00002 Funding details: European Commission, EC Funding text 1: The research was supported by the European Union within the framework of the National Laboratory for Autonomous Systems , project no. RRF-2.3.1-21-2022-00002. The research reported in this paper is part of the project Prin 2020 DigiT-CCAM - Digital Twins per la Mobilità Cooperativa, Connessa e Automatizzata (project no. 2020Z9HEMJ) funded by the Italian Ministry of the University and the Research . AB - Several interdisciplinary studies have investigated the impact of Connected and Autonomous Vehicles (CAVs) on the performance of traffic networks, which expect positive effects. Nevertheless, there will be a transitional period during which both Human-Driven Vehicles (HDVs) and CAVs shall operate simultaneously. Adequate modeling of the interactions between CAVs and HDVs is vital to understand the mixed traffic dynamics. We propose a second-order macroscopic model by reconstructing the backward propagation speed of perturbation based on the dynamic headway distance between vehicles in mixed traffic. The proposed model is validated using microscopic simulations, and it replicates the given traffic scenarios subjected to assorted Penetration Rate (PR) of CAVs. The proposed model is employed to investigate the dynamics of mixed traffic. The results demonstrate that the average traffic velocity and the Level of Service (LOS) significantly improve with the increase in the PR of CAVs. Additionally, the performance of the proposed model is compared with the well-known Jiang-Qing-Zhu (JQZ) model, and it outperforms the JQZ model. The proposed model can be employed in traffic forecasting and real-time traffic control. LA - English DB - MTMT ER - TY - JOUR AU - Kővári, Bálint AU - Angyal, Balint Gergo AU - Bécsi, Tamás TI - Deep Reinforcement Learning combined with RRT for trajectory tracking of autonomous vehicles. JF - TRANSPORTATION RESEARCH PROCEDIA J2 - TRANSP RES PROCEDIA VL - 78 PY - 2024 SP - 246 EP - 253 PG - 8 SN - 2352-1465 DO - 10.1016/j.trpro.2024.02.032 UR - https://m2.mtmt.hu/api/publication/34687657 ID - 34687657 AB - Sample inefficiency is a long-standing problem in Deep Reinforcement Learning based algorithms, which shadows the potential of these techniques. So far, the primary approach for tackling this issue is prioritizing the gathered experiences. However, the strategy behind collecting the experiences received less atention, but it is also a legitimate approach for prioritizing. In this paper, the Rapidly exploring Random Trees algorithm and Deep Reinforcement Learning are combined for the trajectory tracking of autonomous vehicles to mitigate the issues regarding sample efficiency. The core of the concept is to utilize the tremendous explorational power of RRT for covering the state space via experiences for the Agent to diversify its training data buffer. The results demonstrate that this approach outperforms the classic trial-and-error-based concept according to several performance indicators. © 2024 The Authors. Published by ELSEVIER B.V. LA - English DB - MTMT ER - TY - JOUR AU - Fazekas, Zoltán AU - Obaid, Mohammed AU - Gáspár, Péter TI - Choosing Routes in Urban Areas that are Robust Against Minor Nonrecurring Traffic Incidents JF - PERIODICA POLYTECHNICA TRANSPORTATION ENGINEERING J2 - PERIOD POLYTECH TRANSP ENG VL - 52 PY - 2024 IS - 2 SP - 173 EP - 180 PG - 8 SN - 0303-7800 DO - 10.3311/PPtr.23792 UR - https://m2.mtmt.hu/api/publication/34687431 ID - 34687431 N1 - Export Date: 22 March 2024 CODEN: PPTED Correspondence Address: Fazekas, Z.; Institute for Computer Science and Control (HUN-REN SZTAKI), Kende u. 13–17., Hungary; email: fazekas.zoltan@sztaki.hun-ren.hu AB - The paper looks at certain vehicle-level re-routing issues within urban road networks, and related network-level traffic management issues. These arise mostly when the traffic along a route of significance is hindered, slowed down, or even blocked because of some – possibly minor – unexpected, nonrecurring traffic incident at a sensitive road location, or road section. Considerations for planning routes in urban areas – routes that are in some sense robust against such incidents – are presented herein. Also, the on–the–spot detection of traffic queues by an ego-vehicle – relying on data streams from on-board visual line-of-sight (LoS) exteroceptive sensors watching, scanning and monitoring the ego-vehicle's road environment, and by some on-board dedicated real-time detection systems processing and analyzing the incoming data streams – is touched upon. However, this traffic congestion avoidance and mitigating approach – effectuated either by individual autonomous vehicles, or by human drivers – presumes availability of alternative routes, which is not the case for a good portion of the route considered. A route planning approach that could be used for routes with such critical sections is proposed and motivated through an example of an urban route of significance. LA - English DB - MTMT ER - TY - GEN AU - Tettamanti, Tamás AU - Okolie, C.A. AU - Ormándi, Tamás AU - Varga, Balázs TI - A Python tool for SUMO traffic simulation to model start-stop system for comprehensive emission analysis PY - 2024 SP - 1 EP - 21 PG - 21 DO - 10.13140/RG.2.2.12554.24003 UR - https://m2.mtmt.hu/api/publication/34589396 ID - 34589396 LA - English DB - MTMT ER - TY - JOUR AU - Fazekas, Zoltán AU - Obaid, Mohammed AU - Karim, L AU - Gáspár, Péter TI - Urban Traffic Congestion Alleviation Relying on the Vehicles’ On-board Traffic Congestion Detection Capabilities JF - ACTA POLYTECHNICA HUNGARICA J2 - ACTA POLYTECH HUNG VL - 21 PY - 2024 IS - 6 SP - 7 EP - 31 PG - 25 SN - 1785-8860 DO - 10.12700/APH.21.6.2024.6.1 UR - https://m2.mtmt.hu/api/publication/34588733 ID - 34588733 N1 - Funding Agency and Grant Number: European Union within the framework of the National Laboratory for Autonomous Systems [RRF-2.3.1-21-2022-00002]; NRDI Office, Budapest, Hungary; [2018-2.1.10-TETMC-2018-00009] Funding text: This research was supported by the European Union within the framework of the National Laboratory for Autonomous Systems (grant: RRF-2.3.1-21-2022-00002) . It was also supported by the NRDI Office, Budapest, Hungary (grant: 2018-2.1.10-TETMC-2018-00009) . LA - English DB - MTMT ER -