VEHITS 2025 Abstracts

Area 1 - Connected Vehicles

Full Papers

Paper Nr:	103
Title:	Application of Consensus Protocols to Vehicular Communications Scenarios for the Negotiation of Cooperative Traffic Maneuvers
Authors:	Miguel Tavares, Emanuel Vieira, João Almeida and Paulo Bartolomeu
Abstract:	The introduction of Connected and Automated Vehicles (CAVs) has changed the face of the automotive sector, enhancing further developments in cooperative mobility, public safety and improved transportation system management. This paper presents an example study of the application of consensus algorithms in Vehicle-to-Everything (V2X) environments to enable reliable communication among vehicles for the realization of cooperative traffic maneuvers. Among others, an important mechanism employed in this work is the Verifiable Event Extension (VEE), which adds the reliability feature to the V2X communications and ensures trust. In addition to assessing various network conditions in detail, this work analyzes the performance and resiliency trade-offs of different consensus protocols applied to maneuver coordination scenarios, namely Zyzzyva, Prat-ical Byzantine Fault Tolerance (PBFT), HotStuff, and Three-Phase Commit (3PC). The obtained results underline the feasibility of applying robust, highly scalable fault-tolerant solutions to open the way towards a safe deployment of next-generation cooperative and autonomous driving systems.
Download

Short Papers

Paper Nr:	17
Title:	Coupled Lateral and Longitudinal Control for Vehicle Platoons on Curved Roads
Authors:	Bocheng Ma, Yang Zhu and Hongye Su
Abstract:	This paper presents a longitudinal and lateral coupling control model for platoons on roads with varying curvature. The model integrates a 3-DOF vehicle dynamic model with lateral distance errors using the Frenet coordinate system and employs a distributed predictive controller. This approach ensures that vehicles within the platoon maintain ideal inter-vehicle distances, follow the leader’s speed, and remain within their lanes, effectively addressing the “cutting-corner” issue. Joint simulations using Carsim and Matlab demonstrate that the model effectively maintains ideal inter-vehicle distances and tracks the leader’s speed, even with changes in the leader’s velocity.
Download

Paper Nr:	28
Title:	Towards Synergistic Effects of C-ITS Services: Assessing the Joint Impact of GLOSA and CACC on Traffic Efficiency and Sustainability
Authors:	Manuel Walch, Calvin Clausnitzer and Matthias Neubauer
Abstract:	This paper investigates the combined effects of Cooperative Adaptive Cruise Control (CACC) and Green Light Optimal Speed Advisory (GLOSA) on traffic efficiency and sustainability using microscopic traffic simulations. Addressing a gap in the literature, the research focuses on the simultaneous use of these Cooperative Intelligent Transport System (C-ITS) services rather than their individual effects. Simulations were conducted at three test sites with varying traffic characteristics and different penetration rates of C-ITS technologies. The results demonstrate that CACC significantly improves traffic flow and reduces CO2 emissions starting at a 16% penetration rate. However, the effects of GLOSA were marginal and statistically insignificant within the chosen simulation setup. The combined use of CACC and GLOSA provided slight improvements over CACC alone, though these differences were not statistically significant. The findings highlight the substantial benefits of CACC in enhancing traffic flow and reducing emissions, particularly at higher penetration rates. The study underscores the importance of widespread adoption of CACC and calls for further research to explore additional service combinations to optimise the potential of C-ITS for sustainable transportation.
Download

Paper Nr:	29
Title:	Towards Impact Assessment of Cooperative Routing on Traffic Efficiency: A System Dynamics Approach
Authors:	Manuel Walch and Matthias Neubauer
Abstract:	The proliferation of connected vehicles and Cooperative Intelligent Transport Systems (C-ITS) introduces novel opportunities for enhancing various aspects in traffic (e.g., efficiency, sustainability, safety). As C-ITS gains prominence, evaluating its impact requires comprehensive impact assessment studies. While microscopic simulators and Agent-based Models (ABM) dominate C-ITS evaluations, this paper adopts an alternative approach, utilizing System Dynamics (SD) to assess the impact of Cooperative Routing (CR) on traffic efficiency. Thereby a Stock-Flow Model (SFM) is developed, considering parameters such as equipment rates, delay thresholds, and route update intervals. Results indicate that even a low equipment rate (25%) significantly improves traffic efficiency. However, high equipment rates with prolonged route update intervals introduce challenges, causing route overloads and increased delays. These effects are consistent with the current literature on CR using ABM. Furthermore, this study suggests possibilities for model extensions, including predictive rerouting, alternative rerouting criteria, and consideration of sustainability impacts. Overall, these findings contribute to further development in the direction of cooperative connected and automated mobility.
Download

Paper Nr:	67
Title:	Interference-Based Reliability and Capacity Analysis for IEEE 802.11 Broadcast Ad-Hoc Networks on the Highway
Authors:	Zhijuan Li, Xintong Wu, Xiaokun Li and Xiaomin Ma
Abstract:	Interference is a critical factor that degrades wireless network performance. In IEEE 802.11 wireless broadcast networks, hidden terminals and concurrent transmissions are the primary sources of interference due to the carrier sense multiple access (CSMA) mechanism. Therefore, analyzing the signal-to-interference-plusnoise ratio (SINR) distribution is essential for evaluating network performance, whose derivation relates to the transmission probabilities of hidden terminals and concurrent transmissions. In this study, we utilize the existing semi-markov process (SMP) model to calculate these transmission probabilities. Subsequently, we employ the Laplace transform to analyze the SINR distribution in IEEE 802.11 broadcast ad-hoc networks on highways. Based on the derived SINR distribution, we further evaluate the reliability and capacity. This approach can be readily extended to two-dimensional (2D) or three-dimensional (3D) scenarios by employing d-dimensional (1 ≤ d ≤ 3, dD) point process. Experimental results demonstrate that the proposed model achieves high accuracy under small to medium interference ranges. Additionally, the analysis remains highly accurate for receivers within 70 meters, even in scenarios with large interference ranges.
Download

Paper Nr:	14
Title:	Investigating the Safety Effects of Degraded Wireless Performance on Connected Longitudinal Driver Assistance Functions
Authors:	Roland Nagy, Zsombor Pethő, Tamás Márton Kazár, Tibor Turóczi and Árpád Török
Abstract:	In the near future, Vehicle-to-Everything (V2X) based technologies will enable vehicles and other road users to exchange information with each other, even in cases where the applicability of other sensors is limited. This technology will be able to ensure the operation of advanced driver assistance systems, in cases where the other sensors are malfunctioning. In such situations, where only wireless communication can be relied upon, it is essential to be able to react to possible changes in network performance metrics. The objective of this paper is to address the aforementioned problem by characterizing the safety-risk associated with vehicle dynamic parameters and the factors influencing the network performance metrics in different scenarios. The network can be divided into seven distinct layers that are responsible for data transmission, and the research primarily focuses on the physical layer, with the objective of studying its impact on the packets sent. In the research, the Signal-to-Noise Ratio (SNR) is considered to be the primary network influencing parameter. This will facilitate the enhancement of not only the safety of transportation but also its reliability.
Download

Paper Nr:	32
Title:	Adaptive Traffic Management for Emergency Vehicles in Work Zones
Authors:	Fatemeh Bandarian, Saeedeh Ghanadbashi, Abdollah Malekjafarian and Fatemeh Golpayegani
Abstract:	A work zone is a section of road with closed lanes for maintenance, forcing vehicles to merge and creating congestion bottlenecks on the highway. Emergency Vehicles (EVs) are vital for incident response, with response times closely tied to fatality rates. EVs often face challenges when navigating work zones, and despite their importance, little attention has been given to improving their movement through these areas, highlighting the need for a system that enables quicker EV passage in work zones. While traffic management strategies are implemented in work zones, their effectiveness for EVs remains unexplored. This leaves a gap in understanding work zone management for EVs, ensuring their fast and safe passage. This paper proposes the ADAPtive Emergency MERGing (ADAPT-EMERG) algorithm to address this gap. This algorithm controls vehicles longitudinally for smooth merging into the open lane. It integrates merging approaches, headway adjustment between vehicles, and Variable Speed Limit (VSL) rules to set speed limits. Simulation results show that the ADAPT-EMERG algorithm reduces average travel times by 40%, minimises time loss for EVs by an average of 5%, and achieves a throughput increase across various traffic scenarios compared to the state-of-the-art strategies.
Download

Paper Nr:	68
Title:	Misbehavior Detection in Connected Vehicle: Pre-Bayesian Majority Game Framework
Authors:	Adil Attiaoui, Mouna Elmachkour, Abdellatif Kobbane and Marwane Ayaida
Abstract:	Ephemeral networks, such as vehicular ad hoc networks, face significant security challenges due to their transient nature and susceptibility to malicious nodes. Traditional trust mechanisms often struggle with dynamic topologies and short-lived interactions, particularly when adversarial nodes spread misinformation. This paper proposes a dual-game theoretical framework combining pre-Bayesian belief updates with majority voting to enhance collaborative misbehavior detection in decentralized vehicular networks. The approach models node interactions through two sequential games: a pre-Bayesian game where nodes assess information credibility based on individual beliefs, followed by a majority game that aggregates collective decisions to refine trust evaluations. Simulations across scenarios with varying malicious node proportions demonstrate the framework’s adaptability, showing consistent belief convergence toward accurate classifications despite increased adversarial influence. Results indicate robust performance even when 40% of nodes exhibit malicious behavior, though convergence delays highlight challenges in highly adversarial environments. The study underscores the importance of maintaining benign node majorities for system stability and suggests future integrations with machine learning for scalability. This work provides a foundation for secure, real-time decision-making in applications requiring reliable ephemeral networks, such as connected vehicle systems.
Download

Area 2 - Intelligent Transport Systems and Infrastructure

Full Papers

Paper Nr:	35
Title:	Spatio-Temporal Traffic Prediction for Efficient ITS Management
Authors:	Aram Nasser and Vilmos Simon
Abstract:	Traffic forecasting is a crucial element of Intelligent Transportation Systems (ITSs), exerting significant influence on the optimization of urban mobility. Through precise anticipation of traffic patterns, ITS facilitates proactive traffic flow management, leading to a multitude of benefits for both the city and its inhabitants. However, the intricate topological structure of road networks and the changing temporal patterns in traffic create challenging problems that demand solutions considering both the spatial and temporal aspects of traffic characteristics. Most existing traffic prediction models are influenced by Graph Neural Networks (GNNs) to capture the spatial structure of road networks. However, this approach typically relies on the adjacency matrix, which might not always reflect the dynamic state of traffic conditions. In addition, GNNs are not universally applicable across different traffic topologies. What works for one road network may not yield the same results for another, owing to disparities in the number of roads, thus graph nodes, and the unique characteristics of each location. Therefore, in this paper, the Spatio-Temporal Multi-Head Attention (ST-MHA) model is introduced to solve this issue. ST-MHA depends on a modified version of the Multi-Head Attention (MHA) mechanism to capture the spatial structure of the road network implicitly, as well as a GRU-based encoder-decoder structure for integrating the temporal characteristics. Our model outperforms three state-of-the-art baseline models, which include temporal, spatial, and spatio-temporal models. This enhanced performance is evident across three different prediction horizons when evaluated on a real-world traffic dataset.
Download

Paper Nr:	58
Title:	Topology-Aware Prioritized Patching for EV Charging Infrastructure Vulnerabilities
Authors:	Roland Plaka, Mikael Asplund and Simin Nadjm-Tehrani
Abstract:	Modern critical infrastructures are becoming increasingly complex and exposed to cyber-attacks. As with any digitalized system, these systems suffer from vulnerabilities that threaten overall system security. As a result, eliminating vulnerabilities is imperative for security analysts to counteract potential future attacks. However, vulnerability management is time-consuming and expensive because it requires testing, verification, and validation for the patches. Therefore, there is a need to prioritize which vulnerabilities to fix first in an efficient manner. This paper introduces a patching strategy by identifying the attack path that poses the most severe system risk and the patches with the highest potential to mitigate this risk. The risk assessment is based on novel metrics incorporating dynamic exploitability, impact scores, and the network topology. The method is evaluated on a case study based on electric vehicle charging infrastructures. We collect information on vulnerabilities, exploits, and available patches for this domain and instantiate a realistic network model with relevant components, some of which contain vulnerabilities. Our results show that the proposed method outperforms baseline methods to reduce overall system risk.
Download

Paper Nr:	60
Title:	Identification of Traffic Bottlenecks in Central Dhaka Through Spreading Graph-Based Congestion Analysis
Authors:	Manash Sarker, Kazi Sakib and Naushin Nower
Abstract:	The persistent traffic congestion in Dhaka, Bangladesh, calls for innovative and efficient solutions tailored to its unique urban dynamics. This study introduces a novel approach to traffic bottleneck identification that combines congestion levels and their potential to spread, addressing the critical need for targeted traffic management. Our methodology integrates traffic data collection through Google Maps snapshots, congestion intensity mapping, congestion spreading graphs (CSG), maximal spanning trees (MST), and applying Nä ıve Bayes’ theorem to calculate congestion costs. These tools identify bottlenecks by quantifying both congestion impact and propagation costs within the urban road network. Key findings highlight three major bottlenecks: Kawran Bazar, Mohammadpur Bus Stand, and Dhanmondi 32 intersections, validated using the SUMO simulation platform. These points exhibit significant congestion spread and network-wide delays. The proposed methodology not only identifies critical bottlenecks effectively but also offers actionable insights for urban planners and policymakers to devise targeted interventions. This research bridges existing gaps, providing a cost-effective, adaptable framework for mitigating traffic challenges in resource-constrained cities like Dhaka.
Download

Short Papers

Paper Nr:	21
Title:	Machine Learning-Based Anomaly Detection in Smart City Traffic: Performance Comparison and Insights
Authors:	Mohammad Bawaneh and Vilmos Simon
Abstract:	In recent years, urban roads have suffered from substantial traffic congestion due to the rapidly increasing number of road users and vehicles. Some traffic congestion patterns on specific roadways, such as the recurring congestion during morning and evening rush hours, can be foreseen. However, unexpected events, such as incidents, may also cause traffic congestion. Monitoring traffic status poses vital importance for city traffic operators. They can leverage the monitoring system for resource allocation, traffic lights adjusting, and adapting the public transport schedules to alleviate traffic congestion. Machine learning-based methods for anomaly detection are valuable tools for monitoring traffic status and promptly detecting congestion on city roads. In this paper, we comprehensively study the performance of the common machine learning methods for anomaly detection in the traffic congestion detection use case. In addition, we provide methods usage insights based on the study findings by examining the accuracy, detection speed, and computation overhead of the methods to guide the researchers and city operators toward a suitable method based on their needs.
Download

Paper Nr:	23
Title:	A Modular Detection System for Smart Cities: Integrating Monocular and LiDAR Solutions for Scalable Traffic Monitoring
Authors:	Javier Borau-Bernad, Álvaro Ramajo-Ballester, José María Armingol Moreno and Araceli Sanchis de Miguel
Abstract:	As smart cities continue to develop, they require scalable and efficient traffic monitoring systems. This paper presents a modular detection system that switches between monocular and multimodal modes, depending on the available sensors. The monocular mode, based on the MonoLSS algorithm, offers a cost-effective vehicle detection solution using a single camera, ideal for simpler or low-budget setups. In contrast, the multimodal mode integrates camera and LiDAR data via the MVX-Net model, enhancing 3D accuracy in complex traffic scenarios. This dual-mode flexibility allows smart cities to adapt the system to their infrastructure and budgetary needs, ensuring scalability as urban demands evolve. Inference results demonstrate the superior accuracy of the multimodal approach in challenging environments while validating the efficiency of the monocular mode for simpler settings. Therefore, the modular detection system offers a flexible solution that optimizes both cost and performance, effectively addressing the varied requirements of smart city traffic management.
Download

Paper Nr:	33
Title:	T-RAPPI: A Machine Learning Model for the Corredor Metropolitano
Authors:	Deneb Traverso, Gonzalo Pacheco and Pedro Castañeda
Abstract:	The public transportation system in Lima, Peru, faces significant challenges, including bus shortages, long queues, and severe traffic congestion, which diminish service quality. These issues arise from a lack of modern management tools capable of efficiently handling the Metropolitano bus system. This paper introduces TRAPPI, a predictive model based on Random Forest, developed to estimate bus arrival times at Metropolitano stations. Using historical data on bus arrivals and operational parameters, the model achieved exceptional accuracy, with an R² score of 0.9998 and a MAPE of 0.0554%, demonstrating its robustness and ability to minimize prediction errors. The implementation of T-RAPPI represents a substantial improvement over existing systems, providing operators with data-driven insights to optimize route planning and bus allocation. Additionally, the model's integration into the mobile application Metropolitano + enhances the commuting experience by offering users real-time bus arrival predictions, reducing uncertainty and wait times. Future extensions of this work could include incorporating real-time traffic and weather data to further enhance prediction accuracy and expanding the model to other transit systems in Lima and beyond.
Download

Paper Nr:	65
Title:	Bus Arrival Time Prediction via Hybrid LSTM Using GPS-Derived Run and Dwell Times
Authors:	Aigerim Mansurova, Aiganym Mansurova and Aivar Sakhipov
Abstract:	Accurate bus arrival time prediction is essential for improving the reliability and efficiency of public transportation systems. While existing models often rely on complex ensemble architectures or extensive contextual data, this study explores a simplified approach using a hybrid Long Short-Term Memory (LSTM) model. The model processes sequential features, such as stop IDs, run times, and dwell times, through LSTM layers while integrating contextual information, such as trip start hour and day of the week, via dense layers. Comprehensive experiments on GPS data from buses in Kandy, Sri Lanka, demonstrate the model’s superior performance against state-of-the-art baselines. The proposed model achieves a Mean Absolute Error (MAE) of 13.4 seconds, a Mean Absolute Percentage Error (MAPE) of 10.32%, and a Root Mean Square Error (RMSE) of 24.26 seconds, significantly outperforming alternative methods.
Download

Paper Nr:	85
Title:	Multimodal Route Planning Integrating Soft Mobility: A Real-World Case Study for Student Mobility
Authors:	Rekia Abdellaoui, Simon Caillard, Myriam Foucras and David Baudry
Abstract:	Soft and active mobility (SAM) integration into multimodal route planning is a critical innovation for advancing sustainable transportation. This study explores the inclusion of shared (SSAM) and personal (PSAM) soft and active mobility modes within public transport systems. Leveraging a time-expanded model, the proposed approach optimizes route planning by introducing reliability as a novel metric for selecting transportation options. The methodology is tested on real-world data from student commutes in Strasbourg, providing a practical demonstration of its applicability. Results highlight the significant benefits of integrating SSAM and PSAM, including improved route efficiency, enhanced reliability, and seamless transitions within multimodal networks. This case study underlines the potential of combining innovative models with real-world data to address contemporary transportation challenges effectively.
Download

Paper Nr:	97
Title:	A Greedy Dynamic Task Offloading and Service Management in V2V Networks
Authors:	Hanyang Xing
Abstract:	This paper introduces an advanced framework for optimizing task offloading and service caching in Vehicle-to-Vehicle (V2V) communication networks. The proposed approach leverages a greedy algorithm to address key challenges such as offloading latency, energy consumption, and system overhead. By incorporating practical factors such as task size, server storage capacity, and task popularity, the framework efficiently allocates tasks, thereby reducing computational delays and enhancing network performance. The effectiveness of the algorithm is validated through comprehensive simulations that demonstrate significant improvements in both time efficiency and resource utilization compared to existing methodologies. The results underscore the potential for future advancements in V2V networks, particularly in enhancing network stability under high-speed conditions and developing robust communication systems that maximize the use of roadside computational resources.
Download

Paper Nr:	18
Title:	Coordination for Complex Road Conditions at Unsignalized Intersections: A MADDPG Method with Enhanced Data Processing
Authors:	Ruo Chen, Yang Zhu and Hongye Su
Abstract:	In this paper, we use deep reinforcement learning to enable connected and automated vehicles (CAVs) to drive in a intersection with human-driven vehicles. The multi-agent deep deterministic policy gradient (MADDPG) algorithm is improved to be more efficient for data processing, so that it can solve the problem of learning bottlenecks in complex environments, and use sliding control to execute control strategies. Finally, the feasibility of the method is verified in the simulation environment of CARLA.
Download

Paper Nr:	40
Title:	Readiness for Small-Sized Parking Policies: A Path to Sustainable Urban Mobility
Authors:	Paulo Cantillano-Lizana and Giulia Renzi
Abstract:	Urban land use presents critical challenges for cities as they strive to balance stakeholder interests and needs. Several parking management policies have been implemented to optimise space, promoting more sustainable transport modes, such as electric vehicles, and meet specific needs of network users. This study explores the potential for European cities, especially those facing high density, heavy traffic, and limited urban space, to meet the demand for compact vehicles. By examining the European passenger car market, specifically sales of A-segment vehicles, this study establishes the readiness of national markets to adopt such policies. Findings reveal that countries like Malta and Italy, where urban congestion and spatial constraints pose challenges for urban planners, show a high adoption rate of small vehicles. This trend indicates an opportunity to improve urban space efficiency through targeted policies, potentially encouraging a shift towards more compact and fuel-efficient vehicles. By accommodating infrastructure for smaller vehicles, urban planners can support sustainable urban mobility, reduce environmental impacts, and improve accessibility in densely populated areas.
Download

Paper Nr:	57
Title:	Real-Time Cyclist Prioritization with Fuzzy Logic-Based Signal Control
Authors:	Sarah Salem, Pascal Leone and Axel Leonhardt
Abstract:	The integration of cycling into urban traffic systems has increased significantly. Which drives the expansion of dedicated bicycle lanes at intersections to accommodate the growing cyclist volumes while ensuring traffic efficiency and safety. Addressing cyclists’ priority at signalized intersections presents a complex challenge, necessitating tailored traffic signals and control methods. This research proposes a cycling priority strategy for isolated intersections, using fuzzy logic to make high-quality decisions regarding cyclist priority while minimizing delays for all road users. The methodology involves developing a fuzzy logic-based cyclist priority strategy, using input variables such as vehicle queue and cyclist queue to determine cyclist priority. The evaluation, conducted using VISSIM microscopic traffic simulation, demonstrates that the proposed fuzzy logic-based control system effectively reduces delays and stops for cyclists, with an optimal preference threshold (P*) value of 0.7 balancing the needs of both cyclists and motor vehicles. Sensitivity analysis against traditional control methods further emphasises the potential of the fuzzy logic approach to enhance overall traffic efficiency and promote sustainable urban mobility.
Download

Paper Nr:	82
Title:	Evaluation of Detection Approaches for Road Anomalies Based on Accelerometer Data
Authors:	Edvinas Pocevicius, Donatas Paulauskas, Tomas Eglynas, Valdas Jankunas, Sergej Jakovlev, Mindaugas Jusis and Dovydas Lizdenis
Abstract:	Current container security systems record vibrations and shocks, but their potential for creating smart transportation systems remains underutilized. This study analyzes data collected from a truck and discusses a concept for generating road condition maps from accelerometer data. An experiment was conducted by mounting an accelerometer on a container door to gather acceleration data in various transport conditions. The study focuses on analyzing vertical (Z-axis) accelerations as a primary indicator of road anomalies. The developed concept can be integrated into logistics platforms, enabling vehicle drivers and infrastructure managers to respond to road defects in a timely manner.
Download

Area 3 - Intelligent Vehicle Technologies

Full Papers

Paper Nr:	24
Title:	DiverSim: A Customizable Simulation Tool to Generate Diverse Vulnerable Road User Datasets
Authors:	Jon Ander Iñiguez de Gordoa, Martín Hormaetxea, Marcos Nieto, Gorka Vélez and Andoni Mujika
Abstract:	This work presents DiverSim, a highly customizable simulation tool designed for the generation of diverse synthetic datasets of vulnerable road users to address key challenges in pedestrian detection for Advanced Driver Assistance Systems (ADAS). Although recent Deep Learning models have advanced pedestrian detection, their performance still depends on the diversity and inclusivity of training data. DiverSim, developed on Unreal Engine 5, allows users to control various environmental conditions and pedestrian characteristics, including age, gender, ethnicity and mobility aids. The tool features a highly customizable virtual fisheye camera and a Python API for easy configuration and automated data annotation in the ASAM OpenLABEL format. Our experiments demonstrate DiverSim’s capability to evaluate pedestrian detection models across diverse user profiles, revealing potential biases in current state-of-the-art models. By making both the simulator and Python API open source, DiverSim aims to contribute to fairer and more effective AI solutions in the field of transportation safety.
Download

Paper Nr:	25
Title:	Recognition of Typical Highway Driving Scenarios for Intelligent Connected Vehicles Based on Long Short-Term Memory Network
Authors:	Xinjie Feng, Shichun Yang, Zhaoxia Peng, Yuyi Chen, Bin Sun, Jiayi Lu, Rui Wang and Yaoguang Cao
Abstract:	In the complex traffic environment where intelligent connected vehicles (ICVs) and traditional vehicles coexist, accurately identifying the driving scenarios of a vehicle helps ICVs make safer and more efficient decisions, while also enabling performance evaluation across different scenarios to further optimize system capabilities. This paper presents a typical highway driving scenarios recognition model with extensive scenario coverage and high generalizability. The model first categorizes the constituent elements of driving scenarios and extracts the core elements of typical highway scenarios. Then, based on a long short-term memory (LSTM) network architecture, it extracts features from the ego vehicle and surrounding vehicles to identify the typical driving scenarios in which the ego vehicle is located. The model was tested and validated on the HighD dataset, achieving an overall accuracy of 96.74% for four typical highway scenarios: Lane-change, Car-following, Alongside vehicle cut-in, and Preceding vehicle cut-out. Compared to baseline models, the proposed model demonstrated superior performance.
Download

Paper Nr:	45
Title:	A Simulator Study on Car User’s Perceptions in Interaction with Autonomous Shuttles
Authors:	Sagar Meda, Mario Ilic, Tanja Niels and Eftychios Papapanagiotou
Abstract:	This study integrates a driving simulator and microscopic traffic simulation tool to evaluate the impact of autonomous shuttles on driving behavior and safety in a university campus environment. Two scenarios were developed: one featuring a conventional shuttle and another with an autonomous shuttle, allowing a direct comparison of driver perceptions under identical conditions. Results show that perceived safety was higher for conventional shuttles (Mean: 5.909) compared to autonomous shuttles (Mean: 2.818), while stress levels remained consistent across both scenarios. These findings highlight critical human factors and challenges in integrating autonomous shuttles, offering empirical insights into their behavioral and safety implications in mixed-traffic environments.
Download

Paper Nr:	46
Title:	Large Language Model-Informed Geometric Trajectory Embedding for Driving Scenario Retrieval
Authors:	Tin Stribor Sohn, Maximilian Dillitzer, Tim Brühl, Robin Schwager, Tim Dieter Eberhardt, Michael Auerbach and Eric Sax
Abstract:	This paper introduces a Large Language Model-informed geometric embedding for retrieving behavioural driving scenarios from unlabelled trajectory data, aimed at improving the search of real driving data for scenario-based testing. A Variational Recurrent Autoencoder with a Hausdorff Distance-based loss generates trajectory embeddings that capture detailed spatial patterns and interactions, offering enhanced interpretability over traditional mean squared error-based models. The embeddings are further organised through unsupervised clustering using HDBSCAN, grouping scenarios by similarities at the scene, infrastructure, behaviour, and interaction levels. Using GPT-4o for describing scenarios, clusters, and inter-cluster relationships, the approach enables targeted scenario retrieval via a Graph Retrieval-Augmented Generation pipeline, enabling a natural language search of unlabelled trajectories. Evaluation demonstrates a retrieval precision of 80.2% for behavioural queries involving infrastructure, multi-agent interactions, and diverse traffic conditions.

Paper Nr:	52
Title:	A Scenario-Based Simulation Framework for Testing of Highly Automated Railway Systems
Authors:	Michael Wild, Jan Steffen Becker, Carl Schneiders and Eike Möhlmann
Abstract:	Increasing automation is an ongoing effort across various mobility sectors, including the railway domain, promising to address issues such as sustainability, lack of personnel, and enhancing mobility in rural areas. The development of automated railway systems is a challenging task and the validation of safety of such systems in open context remains an open topic. Simulation-based validation of driverless trains can help to ensure safe operation. This paper presents an extension of the open-source train simulator OpenRails to enable doing a closed-loop simulation with the goal of validating the behavior of a system under test within a simulated environment. We propose a possible scenario-based validation approach and present the whole loop including description of an abstract scenario using Traffic Sequence Charts, derivation of a concrete instance of this abstract scenario, and a novel closed-loop play-out. We share our experiences and the current state of our work and give outlook on future directions.

Paper Nr:	55
Title:	Exploring Shared Gaussian Occupancies for Tracking-Free, Scene-Centric Pedestrian Motion Prediction in Autonomous Driving
Authors:	Nico Uhlemann, Melina Wördehoff and Markus Lienkamp
Abstract:	This work introduces a scalable framework for pedestrian motion prediction in urban traffic, tailored for real-world applications in autonomous driving. Existing methods typically predict either individual objects, creating challenges with higher agent counts, or rely on discretized occupancy maps, sacrificing precision. To overcome these limitations, we propose a scene-centric transformer architecture with a cluster-based training approach, capturing pedestrian dynamics through combined probability distributions. This strategy enhances prediction efficiency as groups of nearby agents are unified into a shared representation, thus reducing computational load while still maintaining a continuous output format. Additionally, we investigate a tracking-free design, exploring the feasibility of accurate predictions based solely on object lists without explicit object association. To assess predictive performance, we compare our approach to state-of-the-art trajectory prediction methods, analyzing several metrics while keeping practical applications in mind. Evaluations on a dedicated pedestrian benchmark derived from the Argoverse 2 dataset demonstrate the model’s strong predictive accuracy and highlight the potential for tracking-free future developments.
Download

Paper Nr:	62
Title:	Real-Time Network-Aware Roadside LiDAR Data Compression
Authors:	Md Parvez Mollah, Murugan Sankaradas, Ravi K. Rajendran and Srimat T. Chakradhar
Abstract:	LiDAR technology has emerged as a pivotal tool in Intelligent Transportation Systems (ITS), providing unique capabilities that have significantly transformed roadside traffic applications. However, this transformation comes with a distinct challenge: the immense volume of data generated by LiDAR sensors. These sensors produce vast amounts of data every second, which can overwhelm both private and public 5G networks that are used to connect intersections. This data volume makes it challenging to stream raw sensor data across multiple intersections effectively. This paper proposes an efficient real-time compression method for roadside LiDAR data. Our approach exploits a special characteristic of roadside LiDAR data: the background points are consistent across all frames. We detect these background points and send them to edge servers only once. For each subsequent frame, we filter out the background points and compress only the remaining data. This process achieves significant temporal compression by eliminating redundant background data and substantial spatial compression by focusing only on the filtered points. Our method is sensor-agnostic, exceptionally fast, memory-efficient, and adaptable to varying network conditions. It offers a 2.5x increase in compression rates and improves application-level accuracy by 40% compared to current state-of-the-art methods.
Download

Paper Nr:	69
Title:	FlexCloud: Direct, Modular Georeferencing and Drift-Correction of Point Cloud Maps
Authors:	Maximilian Leitenstern, Marko Alten, Christian Bolea-Schaser, Dominik Kulmer, Marcel Weinmann and Markus Lienkamp
Abstract:	Current software stacks for real-world applications of autonomous driving leverage map information to ensure reliable localization, path planning, and motion prediction. An important field of research is the generation of point cloud maps, referring to the topic of simultaneous localization and mapping (SLAM). As most recent developments do not include global position data, the resulting point cloud maps suffer from internal distortion and missing georeferencing, preventing their use for map-based localization approaches. Therefore, we propose FlexCloud for an automatic georeferencing of point cloud maps created from SLAM. Our approach is designed to work modularly with different SLAM methods, utilizing only the generated local point cloud map and its odometry. Using the corresponding GNSS positions enables direct georeferencing without additional control points. By leveraging a 3D rubber-sheet transformation, we can correct distortions within the map caused by long-term drift while maintaining its structure. Our approach enables the creation of consistent, globally referenced point cloud maps from data collected by a mobile mapping system (MMS). The source code of our work is available at https://github.com/TUMFTM/FlexCloud.
Download

Paper Nr:	79
Title:	OpenLiDARMap: Zero-Drift Point Cloud Mapping Using Map Priors
Authors:	Dominik Kulmer, Maximilian Leitenstern, Marcel Weinmann and Markus Lienkamp
Abstract:	Accurate localization is a critical component of mobile autonomous systems, especially in Global Navigation Satellite Systems (GNSS)-denied environments where traditional methods fail. In such scenarios, environmental sensing is essential for reliable operation. However, approaches such as LiDAR odometry and Simultaneous Localization and Mapping (SLAM) suffer from drift over long distances, especially in the absence of loop closures. Map-based localization offers a robust alternative, but the challenge lies in creating and georeferencing maps without GNSS support. To address this issue, we propose a method for creating georeferenced maps without GNSS by using publicly available data, such as building footprints and surface models derived from sparse aerial scans. Our approach integrates these data with onboard LiDAR scans to produce dense, accurate, georeferenced 3D point cloud maps. By combining an Iterative Closest Point (ICP) scan-to-scan and scan-to-map matching strategy, we achieve high local consistency without suffering from long-term drift. Thus, we eliminate the reliance on GNSS for the creation of georeferenced maps. The results demonstrate that LiDAR-only mapping can produce accurate georeferenced point cloud maps when augmented with existing map priors.
Download

Paper Nr:	96
Title:	Comparison of Point Cloud and Surface Based Mapping for Autonomous Vehicles
Authors:	Christoph Brückner and Lucila Patino-Studencki
Abstract:	Mapping and localization are essential processes in robotics and autonomous systems, providing precise environmental representation and real-time positioning. Unlike Simultaneous Localization and Mapping (SLAM), which combines these tasks simultaneously, mapping and localization are often decoupled in applications that require higher accuracy and efficiency from the outset, like autonomous vehicles. This study summarizes the main families of map representations used in SLAM and investigates the applicability for standalone mapping and localization tasks. Point cloud and surfaced based Mapping Methods, namely KISS-ICP and PUMA are explored and evaluated numerically using the KITTI database. Key performance metrics accuracy, registration time during localization, and map size are analyzed to compare their effectiveness. The results provide insights into the strengths and limitations of SLAM-based techniques when applied to decoupled processes.
Download

Paper Nr:	100
Title:	D-LeDe: A Data Leakage Detection Method for Automotive Perception Systems
Authors:	Md Abu Ahammed Babu, Sushant Kumar Pandey, Darko Durisic, Ashok Chaitanya Koppisetty and Miroslaw Staron
Abstract:	Data leakage is a very common problem that is often overlooked during splitting data into train and test sets before training any ML/DL model. The model performance gets artificially inflated with the presence of data leakage during the evaluation phase which often leads the model to erroneous prediction on real-time deployment. However, detecting the presence of such leakage is challenging, particularly in the object detection context of perception systems where the model needs to be supplied with image data for training. In this study, we conduct a computational experiment to develop a method for detecting data leakage. We then conducted an initial evaluation of the method as a first step on a public dataset, “Kitti”, which is a popular and widely accepted benchmark dataset in the automotive domain. The evaluation results show that our proposed D-LeDe method are able to successfully detect potential data leakage caused by image similarity. A further validation was also provided to justify the evaluation outcome by conducting pair-wise image similarity analysis using perceptual hash (pHash) distance.
Download

Paper Nr:	101
Title:	Vehicle Longitudinal Speed Estimation Using 3DOF Localization Information and Genetic Solver
Authors:	Reza Ghahremaninejad, Kerem Par, Ali Ufuk Peker and Ömer Konan
Abstract:	The accurate vehicle longitudinal speed measurement is vital for many sub-modules of current vehicle control units. Ego velocity estimation in Automated Vehicles (AV) scope is one of the fundamental functionalities required to properly operate many sub-modules like control, planning, and perception. The current speed sensors on most commercial vehicles have different precision and failure rates. To mitigate the faulty behavior of AV modules in vehicle speed sensor failure scenarios, a real-time velocity estimation method can play a redundant role in the vehicle speed sensor. This work attempts to estimate vehicle longitudinal speed having 3DOF real-time localization data. Considering the vehicle dynamic bicycle model, an objective function is formulated, and then a genetic solver solves the single objective optimization problem. The validation of the velocity estimation is discussed by comparing the real-time estimated value with accurate vehicle speed sensor measurement. Results show an acceptable recall of ego longitudinal velocity for redundancy application.

Paper Nr:	102
Title:	Road Signs Perception: Eye Tracking Case Study in Real Road Traffic
Authors:	Kateřina Bucsuházy, Michal Belák, Vendula Gajdůšková and Robert Zůvala
Abstract:	This study investigates driver visual perception of road traffic signs under real road conditions. Using mobile eye tracking technology, we analyzed glance behavior toward various traffic signs and advertisements along urban and highway routes during daytime and nighttime conditions. Results showed significant differences in glance duration and frequency based on sign type, environmental conditions, and the presence of advertisements. Drivers primarily focused on speed limit and directional signs, while advertisements attracted longer glance durations despite their lower frequency of detection. Nighttime conditions generally led to increased glance durations and higher frequencies for most traffic sign types. These findings highlight the importance of optimizing road signage design and placement to improve driver attention and road safety, especially in environments with high visual clutter. Limitations include the exclusion of peripheral vision effects and potential biases introduced by experimental settings.
Download

Short Papers

Paper Nr:	13
Title:	Real-Time Digital Twin for Construction Vehicle Stability Assessment and Visualization with Improved Front-Loader Payload Estimation
Authors:	Théo Tuerlinckx, Sam Weckx, Steven Robyns and Jeroen D. M. De Kooning
Abstract:	The stability assessment of construction vehicles, which are part of a constantly growing market, is of a high importance for safety and working efficiency. For such vehicles, the stability is mainly impacted by the carried payload. In this paper, a state of the art payload estimation method, based on simplified motion equations, is further improved by coupling it with accurate real-time multibody modelling. An example, that allows to reduce the important impact of joints damping on the payload estimation method, is developed and validated in this paper. A reduction of the payload estimation moving window root mean square error from 12.8% to 2.9% is obtained. Finally, the tractor multibody digital-twin is integrated in a real-time system on a physical setup, allowing to process the signals of the tractor and provide an easy to interpret visualization of the vehicle stability to the operator.
Download

Paper Nr:	27
Title:	Analyzing Model Behavior for Driver Emotion Recognition and Drowsiness Detection Using Explainable Artificial Intelligence
Authors:	Diego Caballero García-Alcaide, M. Paz Sesmero, José A. Iglesias and Araceli Sanchis
Abstract:	Traffic accidents, predominantly caused by human error, pose a significant public health threat globally. Despite technological advancements and regulatory measures, the decline in traffic-related fatalities has stagnated. This research addresses the critical need for advanced driver monitoring systems that can accurately detect fatigue and emotional states to mitigate risky driving behaviors. The core of this study lies in developing and evaluating deep learning models, specifically convolutional neural networks (CNN), augmented with explainable artificial intelligence (XAI), for the dual purpose of emotion recognition and drowsiness detection in drivers. By leveraging XAI, we delve into the decision-making processes of our models, offering unprecedented transparency and interpretability in their predictions. Our findings illuminate the intricate interplay between facial expressions and emotional states versus the subtle cues indicative of drowsiness, creating opportunities for more nuanced and effective driver monitoring systems. This work underscores the transfor-mative potential of XAI in fostering trust, refining model behavior, and propelling forward the development of advanced driver assistance systems (ADAS) aimed at enhancing roadway safety and reducing accidents.
Download

Paper Nr:	34
Title:	Adaptive Interaction Field Framework for Risk-Aware Navigation of Driverless Minibus in Pedestrian Zones
Authors:	Qazi Hamza Jan and Karsten Berns
Abstract:	In complex pedestrian zones, the navigation of driverless minibuses faces significant challenges due to varying environmental structures and pedestrian behavior. These zones range from organized pathways to open, unstructured spaces with minimal navigational cues. To address this, dynamic interaction fields are developed around the driverless minibus, adapting in size and shape to real-time movements. To achieve a similar representation as vehicle, interaction fields are developed that incorporate pedestrian unawareness. These virtual fields facilitate safer and more intuitive interactions between vehicles and pedestrians by incorporating real-time pedestrian awareness and activity data. The proposed model assesses risk by aggregating grid values from overlapping zones between pedestrians and driverless minibus, computing potential encounters based on spatial positions and awareness levels. A gradient-based heat map visualizes risk, highlighting areas where interaction with pedestrians is needed. This adaptive approach enables the decision-making module to initiate appropriate responses, such as escape maneuvers or interaction mode activation, based on risk thresholds. The interaction field further classifies risk into ambient, direct, or critical levels, guiding the system’s reactions. This framework enhances safety protocols and situational awareness in diverse urban environments. The vehicle was able to drive and interaction in a better way with enabled interaction fields. Based on these risk values, various interaction modules were activated, facilitating meaningful and context-aware interactions with pedestrians.
Download

Paper Nr:	41
Title:	Integrating 5G into VANETs: Methodological Approaches and Performance Evaluation Through Simulation
Authors:	Nouri Omheni, Mazen Sboui and Faouzi Zarai
Abstract:	The integration of 5G technology with Vehicular Ad-hoc Networks (VANETs) represents a significant advancement in the field of Intelligent Transportation Systems (ITS). This paper explores the potential of 5G to enhance real-time communication between vehicles and infrastructure, aiming to improve road safety and traffic management. Using simulation tools such as OMNeT++ and SUMO, various traffic scenarios were modeled to assess the performance of the integrated 5G-VANET system. Key performance indicators including end to end latency, packet loss, and node acceleration behaviour were evaluated. The results indicate that the integration of 5G reduces communication latency to below 10 milliseconds and achieves packet delivery rates exceeding 95% in high-density traffic situations. This study demonstrates the feasibility of 5G-enhanced VANETs, highlighting their potential to contribute to safer and smarter transportation systems.
Download

Paper Nr:	42
Title:	Enhancing Railway Obstacle Detection System Based on Incremental Learning
Authors:	Qiushi Guo, Bin Cao, Dehao Hao, Cheng Wang, Lijun Chen and Peng Yan
Abstract:	Obstacle detection systems face challenges related to the Catastrophic forgetting problem, where old obstacles may be misclassified when training new unseen obstacles. Re-training a model from scratch for every new obstacle is often impractical. In this work, we propose a continual learning-based approach to efficiently update the model without repeatedly retraining on previous data, while simultaneously mitigating catastrophic forgetting. Experimental results demonstrate the effectiveness of our proposed method.
Download

Paper Nr:	44
Title:	Towards an Optical Chemometric Sensor for Anti-Icing Agents on Asphalt Pavement
Authors:	Benny Thörnberg, Alex Klein-Paste and Wei Zhang
Abstract:	To ensure traffic safety during winter, chemical agents are typically used for de-icing and anti-icing. Smarter and more precise dispersion of chemicals, which considers local variations in concentration, has the potential to reduce the total amount applied. This paper presents a study of an optical chemometric sensor capable of measuring the NaCl concentration and the weight of the dispersed solution per square meter. The experiment was conducted in an indoor environment, where seven solutions of tap water and NaCl were poured onto a diffuse surface made of burned clay. Short-wave infrared light was illuminated onto the surface, and the light was diffusely reflected into a spectrometer after passing through the liquid layer twice. Absorption in the liquid layer alone can be extracted by subtracting the background and further modeled using Beer-Lambert's law. Both the concentration of NaCl and the amount of liquid can be computed by fitting an overdetermined equation system. Experimental results show a strong correlation between actual and computed concentrations, as well as between actual and computed liquid quantities. Suppression of ambient light, spectral variations of asphalt, harsh environments, dynamic range, and signal-to-noise ratio are among the challenges for outdoor chemometric sensing of asphalt pavements.
Download

Paper Nr:	48
Title:	Comparison of Parametrization Approaches for Scenario-Based Testing
Authors:	Christoph Glasmacher, Marcel Sonntag and Lutz Eckstein
Abstract:	Scenario-based testing is a promising approach to assess and assure the safety of automated and connected driving functions. In this approach, test scenarios are often described in an abstract way. Norms sometimes even provide certain parameter values for, e.g., approaching maneuvers in lane-keeping situations. However, the type of parametrization is often not fully specified - neither in databases nor in regulations. This paper assesses differences in possible types of parametrizations for test scenarios and gives guidance about the importance to choose a suitable parametrization for individual use cases. For this, different parametrization types are categorized. The effects on the outcome of tests are investigated in a comprehensive study simulating 435,456 test cases in the CARLA simulator. Thereby, 8 different systems under test are investigated to observe the outcome on different parametrizations on intersections. The results show a high influence of the parametrizations for different systems under test on the test outcomes leading to the need for carefully selecting a suitable parametrization approach.
Download

Paper Nr:	49
Title:	Improving Object Detection Through Multi-Perspective LiDAR Fusion
Authors:	Karl Schrab, Felix Hilgerdenaar, Robert Protzmann and Ilja Radusch
Abstract:	Detection of relevant objects in the driving environment is crucial for autonomous driving. Using LiDAR scans and image detection based on neural networks for this task is one possibility and already well researched. With advances in the V2N communication stack, the task of object detection can be shifted towards the edge-cloud, which would enable collaborative data collection and consideration of multiple perspectives in preparation for the detection. In this paper, we present an initial analysis of this idea, by utilizing the Eclipse MOSAIC co-simulation framework to develop and test the fusion of multi-perspective LiDAR frames and subsequent object detection. We generate synthetic LiDAR data from the views of multiple vehicles for detection training and use them to assess the robustness of our approach in regard to positioning and latency requirements. We found that a data fusion from multiple perspectives primarily improves detection of largely or fully occluded objects, which could help situation recognition and, therefore, decision making.
Download

Paper Nr:	50
Title:	The Components of Collaborative Joint Perception and Prediction: A Conceptual Framework
Authors:	Lei Wan, Hannan Ejaz Keen and Alexey Vinel
Abstract:	Connected Autonomous Vehicles (CAVs) benefit from Vehicle-to-Everything (V2X) communication, which enables the exchange of sensor data to achieve Collaborative Perception (CP). To reduce cumulative errors in perception modules and mitigate the visual occlusion, this paper introduces a new task, Collaborative Joint Perception and Prediction (Co-P&P), and provides a conceptual framework for its implementation to improve motion prediction of surrounding objects, thereby enhancing vehicle awareness in complex traffic scenarios. The framework consists of two decoupled core modules, Collaborative Scene Completion (CSC) and Joint Perception and Prediction (P&P) module, which simplify practical deployment and enhance scalability. Additionally, we outline the challenges in Co-P&P and discuss future directions for this research area.
Download

Paper Nr:	53
Title:	Landmark-Based Geopositioning with Imprecise Map
Authors:	Noël Nadal, Jean-Marc Lasgouttes and Fawzi Nashashibi
Abstract:	This paper introduces a novel approach for real-time global positioning of vehicles, leveraging coarse landmark maps with Gaussian position uncertainty. The proposed method addresses the challenge of precise positioning in complex urban environments, where global navigation satellite system (GNSS) signals alone do not provide sufficient accuracy. Our approach is to achieve a fusion of Gaussian estimates of the vehicle’s current position and orientation, based on observations of the vehicle, and information from the landmark maps. It exploits the Gaussian nature of our data to achieve robust, reliable and efficient positioning, despite the fact that our knowledge of the landmarks may be imprecise and their distribution on the map uneven. It does not rely on any particular type of sensor or vehicle. We have evaluated our method through our custom simulator and verified its effectiveness in obtaining good real-time positional accuracy of the vehicle, even when the GNSS signal is completely lost, even on the scale of a large urban area.
Download

Paper Nr:	56
Title:	Smart Rideshare Matching: Feasibility of Utilizing Personalized Preferences
Authors:	Shaguna Gupta, Shi Feng and B. Brian Park
Abstract:	We investigated the feasibility of utilizing vehicular telematics data for ride-sharing matching. The main focus was to use personalized preferences including home and workplace departure and arrival times. A case study was conducted using the vehicular telematics commuting data between their home to the University of Virginia (UVA) campus. Using data from April 2022, which captures vehicle trips, arrival times, and departure times at UVA, this research analyzed vehicle trips over two weeks to identify individuals with similar commuting schedule preferences. By clustering vehicles based on proximity and timing, we proposed a framework for matching individuals who share similar arrival and departure schedule preferences and live in nearby locations, thereby facilitating coordinated ride-sharing opportunities. The findings are presented through visualizations illustrating ride-matching potential, particularly during peak commuting hours. The matching would offer a convenient ride-sharing solution for UVA commuters while maintaining their commuting flexibility. This approach could also offer a sustainable transportation solution that enhances travel efficiency, lowers environmental impact, and supports the broader adoption of ride-sharing within academic and urban settings. The proposed framework provides a scalable model for systematic ride-sharing implementation and could guide future research and policy development for sustainable campus mobility solutions.

Paper Nr:	61
Title:	Time-Aware Contrastive Representation Learning for Road Network and Trajectory
Authors:	Ashraful Islam Shanto Sikder and Naushin Nower
Abstract:	Modeling and learning representations for road networks and vehicle trajectories is essential for improving various Intelligent Transportation System (ITS) applications. Existing methods often treat road network and trajectory data separately, focus only on one, employ two-step processes that result in information loss and error propagation, or ignore temporal dynamics. To address these limitations, we propose a framework called Time-Aware Contrastive Representation Learning for Road Network and Trajectory (TCRLRT). Our approach introduces an end-to-end model that simultaneously learns road network and trajectory representations, enhanced by a temporal encoding module that captures temporal information and a synthesized hard negative sampling module to enhance the discriminative power of the learned representations. We validate the effectiveness of TCRLRT through extensive experiments conducted on two real-world datasets, demonstrating improved performance over baseline methods across multiple downstream tasks. The results highlight the advantages of joint representation learning with temporal modeling and hard negative sampling, leading to robust and versatile representations.
Download

Paper Nr:	75
Title:	Predicting Depth Maps from Single RGB Images and Addressing Missing Information in Depth Estimation
Authors:	Mohamad Mofeed Chaar, Jamal Raiyn and Galia Weidl
Abstract:	Depth imaging is a crucial area in Autonomous Driving Systems (ADS), as it plays a key role in detecting and measuring objects in the vehicle’s surroundings. However, a significant challenge in this domain arises from missing information in Depth images, where certain points are not measurable due to gaps or inconsistencies in pixel data. Our research addresses two key tasks to overcome this challenge. First, we developed an algorithm using a multi-layered training approach to generate Depth images from a single RGB image. Second, we addressed the issue of missing information in Depth images by applying our algorithm to rectify these gaps, resulting in Depth images with complete and accurate data. We further tested our algorithm on the Cityscapes dataset and successfully resolved the missing information in its Depth images, demonstrating the effectiveness of our approach in real-world urban environments.
Download

Paper Nr:	78
Title:	ADORe: Unified Modular Framework for Vehicle and Infrastructure-Based System Level Automation
Authors:	Mikkel Skov Maarssoe, Sanath Konthala, Marko Mizdrak, Giovanni Lucente, Matthias Nichting, Thomas Lobig and Andrew Koerner
Abstract:	Recent advancements in automated driving have primarily focused on achieving autonomy within individual vehicles. However, a broader paradigm shift is emerging that leverages both vehicle-level autonomy and collaboration with other road users and infrastructure to optimize traffic flow and enhance safety. This paper introduces ADORe®(Automated Driving Open Research), an open-source Automated Driving System developed by the German Aerospace Center (DLR). ADORe adopts a modular, system-level approach, enabling seamless integration between Single-Agent Automated Driving for local autonomy and Multi-Agent Autonomous Driving for infrastructure-assisted decision-making. By utilizing vehicle-to-infrastructure (V2X) communication, ADORe facilitates coordinated multi-agent planning, dynamic route optimization, and improved situational awareness through shared data. The framework supports flexible testing via simulation tools like CARLA and SUMO, alongside deployment on research vehicles equipped with advanced sensors and teleoperation capabilities. Successful demonstrations in research projects like the German national Gaia-X4ROMS and MAD Urban validate ADORe’s capability to bridge the gap between isolated autonomous driving and cooperative traffic systems. This collaborative approach highlights the potential of automated driving systems as a cornerstone of intelligent transportation systems, advancing safety, efficiency, and interoperability.

Paper Nr:	88
Title:	A Framework for Disentangling Efficiency from Effectiveness in External HMI Evaluation Procedures for Automated Vehicles
Authors:	Alexandros Rouchitsas
Abstract:	Automated vehicles (AVs) are rapidly transforming smart cities, offering potential benefits such as improved safety, performance, mobility, accessibility, and overall user experience in traffic. A key area of focus in this evolution is the development of external human-machine interfaces (eHMIs) which aim to equip AVs with communication capabilities. Said interfaces address critical challenges, including mitigating safety risks and enhancing traffic flow in scenarios where drivers are inattentive or altogether absent, and play an important role in allaying distrust of the general public in AVs. Considering the research field of eHMIs is relatively young, it is unsurprising that standardized eHMI evaluation procedures are yet to be established. As a result, the effectiveness and efficiency of eHMI concepts are often assessed either simultaneously within the same evaluation procedure or separately but in otherwise similar procedures. Unfortunately, these approaches overlook on the whole the fundamental differences between the two constructs, resulting in limitations relating to the validity, reliability, and comparability of the findings. Here, I present a definitive framework aimed at disentangling efficiency from effectiveness by guiding methodological choices regarding design rationale explanation, instructions emphasizing speed, trial-level time limit, and targeted performance measures, depending on the research questions of interest.
Download

Paper Nr:	89
Title:	Enhancing Scenario-based Testing for Automated Driving Systems: An Ontology-Based Scenario Modeling Framework
Authors:	Zhenguo Cui, Svetlana Dicheva, Adam Abdin, Bernard Yannou and Jean-Marc Giroux
Abstract:	Scenario-based Testing (SbT) emerges as a pivotal approach for validating the safe behaviors of Automated Driving (AD) and Advanced Driver-Assistance Systems (ADAS). Using virtual simulation, SbT allows for generating and running massive testing cases. This approach gathers typical driving situations and critical edge cases. Properly modeling representative scenarios is a primary challenge. A scenario model needs to account for complex components, such as roads, infrastructure, road users, and their behaviors and interactions. Ontology-based frameworks are proposed to model scenarios in a detailed manner. However, some limitations exist, such as (i) expressing dynamic behaviors, (ii) the capacity in complex scenario modeling to achieve more realistic simulation; and (iii) ensuring comprehensive ontology coverage and plausibility. This paper proposes an ontology framework addressing these shortcomings. A comparative evaluation is conducted using the developed quantitative metrics to assess the ontology framework against two other industrial ontologies.
Download

Paper Nr:	92
Title:	Quality and Trust Indicators of Digital Road Infrastructure Data Are Essential to Improve Its Usability: An Intelligent Speed Assist (ISA) Study
Authors:	Jacco van de Sluis, Daniel Altgassen and Peter-Paul Schackmann
Abstract:	The promise of a Digital Road Infrastructure (DRI) is to improve both road and vehicle safety. ADAS/ADS with DRI support, can help vehicles in overcoming certain sensor limitations, handle more complex operational situations and offer additional situational awareness. An effective DRI must be trusted and must offer the required data quality, both are currently lacking. Intelligent Speed Assist (ISA) is an interesting show case for the added value of DRI. In our approach camera-based traffic sign detections and map-based speed limit information, both occasionally wrong, are augmented with actual speed limit and road layout information coming from DRI. A Misbehaviour Detection and Reporting (MBD&R) concept tailored to the ISA sources is deployed in the vehicle to detect and report ISA related misbehaviour. Trust and quality indicators are calculated for data coming from camera, map and DRI, which are used to verify and compare theses sources and make improved ISA speed limit decisions. The vehicle implementation is tested under real-life traffic conditions. Our work is a first step in realizing a trusted DRI. The long-term goal is collaboration among all stakeholders to implement mechanisms that improve trust and the quality of shared data sources for use in traffic safety applications.
Download

Paper Nr:	94
Title:	A Dynamic Traffic Management System for Itinerary Optimization
Authors:	Iván Monzón Catalán, Vicente R. Tomás López and Lú ıs A. Garć ıa Fernández
Abstract:	Freight transport is a fundamental pillar of the European economy, accounting for more than 9 % of the European Union’s Gross Domestic Product (GDP). Seaports are critical nodes in this logistics chain, handling more than 67.8% of freight transport in tonne-kilometres. This article presents an intelligent traffic management system designed specifically for road access to the Port of Rotterdam, the largest and busiest port in Europe, with the aim of optimising the flow of vehicles, improving operational efficiency and reducing CO2 emissions.
Download

Paper Nr:	12
Title:	A Universal Railway Obstacle Detection System Based on Optical-Flow Guided Semi-Supervised Segmentation
Authors:	Qiushi Guo, Bin Cao, Dehao Hao, Cheng Wang, Lijun Chen and Peng Yan
Abstract:	Detecting obstacles in railway scenarios is both crucial and challenging due to the wide range of obstacle categories and varying ambient conditions such as weather and light. Given the impossibility of encompassing all obstacle categories during the training stage, we address this out-of-distribution (OOD) issue with a semi-supervised segmentation approach guided by optical flow clues. We reformulate the task as a binary segmentation problem instead of the traditional object detection approach. To mitigate data shortages, we generate highly realistic synthetic images using Segment Anything (SAM) and YOLO, eliminating the need for manual annotation to produce abundant pixel-level annotations. Additionally, we leverage optical flow as prior knowledge to train the model effectively. Several experiments are conducted, demonstrating the feasibility and effectiveness of our approach.
Download

Paper Nr:	20
Title:	A Two-Stage Extended Kalman Filter-Based Approach Against FDI Cyber-Attack in Intelligent and Connected Vehicles
Authors:	Bin Sun, Shichun Yang, Yu Wang, Jiayi Lu and Yaoguang Cao
Abstract:	With the widespread integration of artificial intelligence and telecommunication technologies in vehicles, the challenge of cybersecurity in Intelligent and Connected Vehicles (ICVs) has gained significant attention. A typical and high-risk cyber-attack technique involves False Data Injection (FDI) into sensors through the network, resulting in deviations in subsequent planning and control algorithm outcomes. Existing approaches suffer from limited robustness, being suitable only for simple models or requiring extensive data for the training model, which limits their practicality. Therefore, this paper proposes a method based on a Two-stage Extended Kalman Filter (TSEKF), which not only detects cyber-attacks but also restores the vehicle’s true motion state, thereby enhancing the robustness of vehicle ego state perception. The experimental results demonstrate that the proposed method exhibits strong performance across various motion scenarios, offering an effective solution for the safe operation of ICVs.
Download

Paper Nr:	37
Title:	Technology Acceptance Modelling for Investigating the Uptake of Electric, Connected, Autonomous, and Shared Mobility Technologies
Authors:	Konstantina Karathansopoulou, Despoina Mitsiogianni, Eleni Tsaousi, George Dimitrakopoulos and Dimitris Georgiadis
Abstract:	This paper introduces QETAM (Quantitative Effect and Technology Acceptance Modelling), the first quantitative user acceptance model for evaluating the impact and adoption of Electric, Connected, Autonomous, and Shared (ECAS) mobility technologies. Developed in Python, QETAM leverages data collected through specifically designed questionnaires to assess key adoption factors, including technological reliability, user attitudes, infrastructure, and environmental considerations. The model accounts for the interconnected nature of ECAS technologies, emphasizing synergies between electric propulsion, connectivity, autonomy, and shared mobility services. Utilizing advanced statistical techniques, it analyzes large-scale datasets to provide a data-driven understanding of user behavior. Beyond academic contributions, QETAM offers practical insights for policymakers and industry stakeholders, supporting the transition toward sustainable and user-centric mobility solutions.
Download

Paper Nr:	77
Title:	Anomaly Detection for Traffic Management Purposes at Urban Intersections Using Infrastructure-Generated Vehicle-to-X Messages
Authors:	Ina Partzsch, Adrien Bellanger, Michael Klöppel-Gersdorf, Rutuja Mohekar, Friedrich Trauzettel and Thomas Otto
Abstract:	Reliable detection of problematic system states in traffic management poses a significant challenge. Failures in detection can result in the inability to intervene in a timely manner, while excessive detection may lead to operator fatigue, causing critical information to be ignored amidst an overload of irrelevant messages. Light-controlled intersections represent both safety and efficiency-critical locations within urban traffic networks. Anomalies in these traffic system units can manifest at various levels: technically/physically within the control systems (actuators, sensors, communication technology), at the traffic data level (reliability and completeness of collected traffic data), and in traffic observation (unusual traffic flows, unusual objects). Anomaly detection occurs across these different levels using various methods (technical and algorithmic). Vehicle-to-Everything (V2X) communication provides an additional data source for monitoring the correct and efficient operation of traffic signal systems. This paper presents strategies for leveraging the diverse messages from V2X communication to identify unusual system states across these levels. We demonstrate our approaches at an urban intersection within the Digital Testbed Dresden.
Download

Paper Nr:	81
Title:	A Case Study on Defining Infrastructure Sensor Positions with Consideration of Existing Infrastructure
Authors:	Philipp Klein
Abstract:	Support by infrastructure sensors can be crucial to enable automated vehicles to safely navigate complex urban driving environments. Finding the suitable positions for infrastructure sensors is a complex problem with different demands and factors. This paper proposes a method of automating the process of selecting positions for infrastructure sensors in a 2D environment. The positions are selected using available data of the streets, for sensor placement suitable existing infrastructure and sensor coverage demands. This methodology is then applied to finding sensor positions in the neighborhood of Lausitzer Platz in Berlin, Germany. The sensor demands for this are to taken from a virtual roll out scenario of the U-Shift vehicle concept. This is done by first finding suitable sensor positions for the bigger streets with the highest cargo and person transportation demand and then covering of every street in the neighborhood. In this use case more than half the sensor could be placed on existing infrastructure, if there is a high density of existing infrastructure that is suitable for the placement of sensors.
Download

Paper Nr:	93
Title:	A Tool for V2X Infrastructure Placement
Authors:	Michael Klöppel-Gersdorf and Joerg Holfeld
Abstract:	V2X technology sees an increasing rollout all over Europe, for instance as part of the C-Roads initiative. These rollouts are put into operation with implementing several use cases like Traffic Signal Priority request (TSP) for public transport or emergency vehicles or the provision of Green-Light Optimized Speed Advisory (GLOSA). Depending on the use case, the placement of V2X communication equipment, like Road-Side Units (RSUs), is essential for successful implementation of services. In this paper, a tool for V2X planning is introduced, which allows the efficient and fast estimation of V2X communication ranges especially in densely developed areas, reducing the need for costly measurement campaigns. Predicted data is compared with the results of a real-world measurement campaign in the city of Chemnitz, Germany.
Download

Area 4 - Data Analytics

Full Papers

Paper Nr:	66
Title:	Intersection with Highly Adaptive Traffic Lights Can Still Be Suitable for C-ITS Service GLOSA
Authors:	Thomas Otto, Ina Partzsch and Michael Klöppel-Gersdorf
Abstract:	The C-ITS service “Green Light Optimal Speed Advisory” (GLOSA) is a highly promising Day1.0 services for efficient, environmentally friendly and safe cooperative transport. It will play a crucial role in the increasing automation of assistance and driving functions in cooperative connected automated mobility (CCAM). Over the past three decades, significant efforts have been made to make traffic light signalling as adaptable as possible to traffic requests. However, there is a common belief that the more flexible the control, the less reliable the forecast and thus the functionality of GLOSA. This paper introduces stability indicators to demonstrate that this belief is only partially accurate. The proposed tool allows for the analysis of historical data from existing systems to derive an indicator for the quality and suitability of the C-ITS GLOSA application. We demonstrate the feasibility of the approach using real world data from the C-ITS corridor Hamburg, Germany.
Download

Paper Nr:	72
Title:	Energy Consumption Prediction with Uncertainty Quantification for Electric Truck Operations: A Data-Driven Approach
Authors:	Rik Litjens, Róbinson Medina, Nikos Avramis, Camiel Beckers, Steven Wilkins and Mykola Pechenizkiy
Abstract:	The adoption of electric trucks in commercial applications is growing due the the adoption of zero-emission zones in large cities. However, the usage of these trucks shows challenges for fleet managers due to their limited range and uncertain energy usage. Accurately predicting the energy consumption of these vehicles is crucial for their optimal usage in commercial applications. This work introduces a novel energy consumption prediction method for electric trucks, based on a data-driven approach. The approach uses a two-stage Long Short-Term Memory (LSTM) architecture: the first stage predicts vehicle speed while the second predicts energy consumption. For the second stage, two updates to the LSTM encoder are proposed. The first improves the energy prediction by splitting the predictions into regenerated and consumed energy, whereas the second provides a score that quantifies the prediction uncertainty using Student’s t-distribution. Evaluating the approach using real-world truck-operation data shows that splitting the energy consumption into regenerative and consumed components contributes to a 20% reduction of error compared to a state-of-the-art LSTM model, mainly due to improved prediction accuracy for regenerated energy. Finally, the t-score demonstrates a 92% reduction of calibration error compared to a Gaussian equivalent. This ensures reliable application in the design of worst-case planning scenarios, decision thresholds, and probabilistic planning approaches.
Download

Paper Nr:	111
Title:	Towards Assessing Cycleway Pavement Surface Roughness Using an Action Camera with IMU and GPS
Authors:	Muhammad Hassam Baig, Jeziel Antonio Ayala Garcia, Waqar Shahid Qureshi and Ihsan Ullah
Abstract:	This paper introduces an autonomous and cost-effective method for assessing cycleway pavement roughness, using an action camera equipped with high-resolution sensors including an Inertial Measurement Unit (IMU) and a Global Positioning System (GPS). The methodology utilizes simplified quarter car model for bicycles, without manual intervention, to calculate International Roughness Index (IRI) for cycleway surface quality evaluation. It utilizes our novel approach to determine stable section from which average acceleration orientation vector is computed. For analysis we propose a corrected-roughness index (CRI), which is a quantized version of IRI. Experiments conducted on asphalt cycleways in Ireland revealed strong correlations between vehicle vibration and surface roughness. Results further demonstrate the consistency of the proposed model across different bikes through comparative analysis. Observations indicate bias in vibration data, influenced by different tire sizes and the mechanical features of the bicycles.
Download

Short Papers

Paper Nr:	43
Title:	Hybrid Framework for Real-Time Traffic Flow Estimation Using Breadth-First Search
Authors:	Sajjad Mahdaviabbasabad, Ynte Vanderhoydonc, Roeland Vandenberghe and Siegfried Mercelis
Abstract:	Traffic flow data is essential for urban planning, logistics, transport management, and similar applications. However, achieving full sensor coverage across a road network is often infeasible due to high installation and maintenance costs. Simulation data from traffic models can help in filling this gap. However, calibrating and validating these traffic models is time-consuming. This paper presents a framework that combines real-time traffic flow predictions from sensor-equipped road segments with 24-hour static simulation data across an entire network. By applying a method based on the Breadth-First Search algorithm, this framework updates network-wide traffic flow by utilizing the data-driven predictions at sensor-equipped road segments and simulation data. Evaluation on a network with over 27000 road segments shows that this approach improves prediction accuracy over static simulation and is viable for real-time deployment.
Download

Paper Nr:	112
Title:	Intelligent Pavement Condition Rating System for Cycle Routes and Greenways
Authors:	Syed M. Haider Shah, Waqar Shahid Qureshi, Gerard O’ Dea and Ihsan Ullah
Abstract:	This study introduces an intelligent framework for assessing cycling infrastructure, addressing the limitations of traditional pavement evaluation methods. At the core of the system is the CRSI, a 1-to-5 rating scale specifically designed to evaluate cycle routes based on critical factors like surface quality, vegetation encroachment, and drainage. A dataset of over 40,000 frames, extracted from videos captured using handlebar-mounted GoPro cameras and annotated by experts, forms the foundation of the system. Four deep learning (DL) models LeNet, AlexNet, EfficientNet-B2, and Swin Transformer-Tiny were trained and evaluated for Cycle Route Surface Index (CRSI) classification. Among all models, Swin Transformer-Tiny performed the best, achieving an impressive accuracy of 99.90%. To further test its robustness, we evaluated the system on four new videos, from which four separate frame sets were generated. Among these, Swin Transformer-Tiny again delivered the highest accuracy, reaching 86.67%, confirming its reliability across different datasets. This CRSI-based framework provides a scalable, automated solution for evaluating cycling infrastructure, empowering transportation agencies to improve maintenance and ensure safer, more accessible cycling networks.
Download

Paper Nr:	22
Title:	Unraveling Urban Traffic Congestion Patterns in Bangladesh
Authors:	Md. Babul Hasan and Manash Sarker
Abstract:	This research presents a comprehensive study on divisional traffic analysis and clustering in Bangladesh, leveraging Google Maps and image processing techniques for traffic intensity data collection across all divisions from January 2023 to June 2023. A total of 1,39,008 snapshots were captured at 15-minute intervals, yielding a detailed traffic dataset. We conducted an in-depth analysis of the collected time series data, focusing on its decomposition into trend, seasonal, and random components (Y = T * S * R). To enhance clustering accuracy, we proposed a modification technique by dividing traffic intensity (Y) by the random fluctuations (R) to minimize random noise in the data preprocessing stage. We implemented Modified Hierarchical Clustering with Dynamic Time Warping (DTW) for clustering, demonstrating superior similarities-pattern extraction compared to traditional hierarchical clustering. Our results identified four distinct traffic clusters. This study provides insights into regional traffic behaviors and offers a robust approach to clustering traffic data, contributing to Bangladesh’s more effective traffic management strategies.
Download

Paper Nr:	113
Title:	Enhancing Pavement Condition Assessment: A Comprehensive Review of Affordable Sensing Technologies for Cycle Tracks
Authors:	Jeziel Antonio Ayala Garcia, Syed Shah, Muhammad Hassam Baig, Waqar Shahid Qureshi and Ihsan Ullah
Abstract:	This paper explores the potential of low-cost sensing technologies for assessing the condition of cycling track pavement. As cycling gains popularity, the demand for efficient pavement maintenance solutions increases. Machine survey-based pavement condition assessment often relies on expensive, specialized equipment, which is not always suitable for cycle tracks due to limited budgets and accessibility, hence the need for low-cost solutions. The integration of low-cost sensing technologies and data collection, such as inertial measurement units (IMUs), low-cost imaging sensors, and crowdsourced data collection, presents a promising alternative for cycle track pavement surface assessment. This paper highlights the advantages of employing out-of-the-shelf devices such as smartphones with GPS, cameras, and IMU, or action cameras with GPS and IMU to collect images, their location, and vehicle vibration in a specific orientation. This data is then used to estimate pavement roughness, detect surface distress, and compute pavement surface condition. Literature reviews reveal a significant gap in the utilization of these technologies for cycle tracks, suggesting a promising area for further research and application. Furthermore, it proposes a software framework for data collection and visualization to combine these technologies to enhance the efficiency and reliability of pavement assessment for cycling infrastructure at a lower cost than machine-based pavement assessment and faster and more quantitative than manual surveys. This paper emphasizes the need for more practical and scalable solutions that support the maintenance of sustainable transportation systems.
Download

Area 5 - Smart Mobility and Sustainable Transport Services

Full Papers

Paper Nr:	30
Title:	Development of a New Architecture for next Generation e-Bikes
Authors:	Tiago Gândara and José Santos
Abstract:	The growing e-bike market demands more efficient, connected, and user-friendly systems. However, existing e-bike architectures are closed, limiting the integration of new technologies such as power management algorithms and security features. This paper proposes a new system architecture utilizing a microcontroller-based motor controller and CAN bus, allowing integration and data exchange with external devices. Experimental testing was conducted to validate the system’s functionality, including testing energy efficiency improvements and security features such as emergency stop protocols. Results demonstrate that the proposed architecture can enhance energy efficiency and provides reliable security, offering a flexible and scalable solution for future e-bike developments.
Download

Paper Nr:	47
Title:	Routing and Charge Planning Strategies for Ridesharing EV Fleets
Authors:	Ashutosh Singh and Arobinda Gupta
Abstract:	Ridesharing systems have become an important part of urban transportation. At the same time, electric vehicle (EV) adoption is also growing at a fast pace as an eco-friendly and sustainable transportation option. To operate a ridesharing system with EV fleets, scheduling an EV fleet to serve passenger requests requires consideration of both the requests, and the available charge and potential future charge requirements of the EVs. In this paper, we address the problem of scheduling EVs by a ridesharing operator, and propose four algorithms that schedule passenger requests while taking into consideration charging requirements of the EVs. Detailed simulation results are presented on a real world data set to show that the algorithms perform well.
Download

Paper Nr:	80
Title:	Understanding Car Usage Patterns for V2G Integration: Insights from Dutch Travel Diaries
Authors:	Simon Leu, Gonçalo Homem de Almeida Correia, Hans van Lint and Axel Leonhardt
Abstract:	Integrating renewable energy sources, such as solar and wind, challenges grid stability due to their intermittent nature. Vehicle-to-grid (V2G) technology provides a promising solution by utilizing electric vehicles (EVs) as decentralized energy storage systems, enabling the storage of surplus energy during low demand and its release during peak demand. The effectiveness of V2G depends critically on car usage patterns. Data from the Netherlands Mobility Panel (MPN) of 2022, comprising travel diaries from 2,505 households, was analyzed to explore this. A methodology was developed to create car usage profiles based on parking durations and locations, distinguishing weekday and weekend patterns. The analysis shows that vehicles are predominantly parked at home, with weekday profiles reflecting work-related parking and weekend profiles highlighting increased leisure activity. Households with shared cars showed higher driving activity and shorter parking durations than households with a 1:1 car-to-license ratio or surplus vehicles. Six distinct car usage clusters were identified for weekdays and four for weekends.
Download

Short Papers

Paper Nr:	31
Title:	Development and Validation of a Vehicle Corner Test Rig Designed for Hardware-in-the-Loop Testing
Authors:	Paulius Kojis, Viktor Skrickij, Valentin Ivanov, Eldar Šabanovič, Marijonas Bogdevičius and Tomas Grikenis
Abstract:	Many original equipment manufacturers and Tier 1 suppliers have recently proposed various vehicle corner concepts. This technology offers numerous benefits for enhancing vehicle dynamics but presents several challenges. Field testing of the corners installed on a vehicle demonstrator could be an ideal solution to address these challenges. Nevertheless, the proof-of-concept development phase also requires studies on the component and system levels. In this regard, a vehicle corner test rig designed for hardware-in-the-loop testing provides a balanced alternative that combines the accuracy, complexity, and accessibility of experimental works. Real suspension components, including a rotating wheel with tyre, provide realistic suspension kinematics and compliance that are very similar to those experienced in real driving scenarios. However, this approach has limitations because the tyre contact forces and loading conditions cannot be fully replicated in a laboratory environment. This paper explores these aspects and describes a developed comprehensive methodology for eliminating inaccuracies, with results validated accordingly.
Download

Paper Nr:	51
Title:	Dynamic Charging on the Go: Optimizing Mobile Charging Stations for Electric Vehicle Infrastructure
Authors:	Suhas Jain and Arobinda Gupta
Abstract:	With the growing adoption of electric vehicles (EVs) as an eco-friendly and sustainable means of transportation, availability of adequate EV charging infrastructure has become very important. While fixed charging stations are the primary means for public charging, they can be effectively augmented by mobile charging stations. In this paper, we address the problem of planning and operating a mobile charging station fleet by an operator. We propose an algorithm for the operator to use for route planning of the MCSs and for scheduling charging requests of EVs to them to try to maximize the number of charging requests served. Detailed simulation results are presented in different realistic scenarios to show that the proposed algorithm works well.
Download

Paper Nr:	64
Title:	Integration of On-Demand Ride-Pooling into a Microscopic Traffic Flow Simulation Environment
Authors:	Oytun Arslan and Silja Hoffmann
Abstract:	The aim of this study is to model an on-demand ride-pooling system in a microscopic traffic flow simulation environment. On-demand systems are hardly modeled in microsimulations, hence a research gap was identified here. First of all, a methodology was developed to match the requests with the vehicles, considering the operator as well as the passenger interests. Then the simulation setup for the campus of the University of the Bundeswehr Munich was carried out, taking into account the specialties for on-demand traffic and the traffic data feed for the remaining traffic. The data flow from already installed traffic cameras into the microsimulation was illustrated and the steps to model on-demand traffic were elaborated. Results have shown that the objective function can be modified according to the ride-pooling system requirements via various weighting parametrizations, that higher demand profiles lead to more shareability and efficiency, and that ride-pooling strategy outperforms ride-hailing in on-demand systems.
Download

Paper Nr:	70
Title:	Impact of Fleet Electrification and Charging Infrastructure on Free-Floating Car Sharing in Milan
Authors:	Sofia Borgosano, Alessandro Nocera, Michela Longo and Wahiba Yaici
Abstract:	The automotive industry’s transition toward sustainability has prioritized Electric Vehicles (EVs) due to their potential to reduce pollution and improve energy efficiency. This evolution is particularly critical in urban contexts such as Milan, where free-floating car sharing services present unique challenges and opportunities for electrification. The integration of EVs into car sharing fleets demands careful consideration of battery autonomy, charging times, and the distribution of charging infrastructure to meet high vehicle utilization rates. This study evaluates the feasibility of transitioning Milan’s internal combustion car-sharing fleet to an electric model, analyzing technical and operational challenges through a scenario-based simulation approach.
Download

Paper Nr:	74
Title:	Data-Driven Analysis of Bicycle Lane Safety in Mexican Cities: Towards a Real-Time Route Recommendation System for Cyclists
Authors:	Carlos Alberto Domínguez-Báez, Ricardo Mendoza-González and Huizilopoztli Luna-García
Abstract:	This study initiated a project to identify urban cycling routes with a focus on cyclist safety in Mexican cities. A Data-Driven Analysis approach was implemented to map the riskiest and safest cycling routes by analysing traffic accident data from national, state, and local datasets. The accident hotspots were visually integrated into the urban map of Guadalajara city (Jalisco, Mexico), to identify high-risk zones for cyclists. The integration of diverse data sources and geospatial analysis allowed for an accurate characterization of accident patterns, providing a clear identification of critical areas. Key results from this initial stage of the project included an accurate risk-zones identification, a replicable methodology for data integration, and a first approach to developing algorithms for cyclist accident analysis. These preliminary findings hold promise for enhancing urban cycling safety and supporting urban eco-mobility strategies in Mexican cities. Additionally, the results served as a foundation for future exploration of machine learning techniques to refine data processing and develop a real-time safe bicycle lane recommender prototype aimed at guiding cyclists toward safer alternatives.
Download

Paper Nr:	76
Title:	RUTGe: Realistic Urban Traffic Generator for Urban Environments Using Deep Reinforcement Learning and SUMO Simulator
Authors:	Alberto Bazán Guillén, Pablo A. Barbecho Bautista and Mónica Aguilar Igartua
Abstract:	We are witnessing a profound shift in societal and political attitudes, driven by the visible consequences of climate change in urban environments. Urban planners, public transport providers, and traffic managers are urgently reimagining cities to promote sustainable mobility and expand green spaces for pedestrians, bicycles, and scooters. To design more sustainable cities, urban planners require realistic simulation tools to optimize mobility, identify location for car chargers, convert streets to pedestrian zones, and evaluate the impact of alternative configurations. However, realistic traffic profiles are essential to produce meaningful simulation results. Addressing this need, we propose a traffic generator based on deep reinforcement learning integrated with the SUMO simulator. This tool learns to generate an instantaneous number of vehicles throughout the day, aligning closely with the target profiles observed at the traffic monitoring stations. Our approach generates accurate 24-hour traffic patterns for any city using minimal statistical data, achieving higher accuracy compared to existing alternatives. In particular, our proposal demonstrates a highly accurate 24-hour traffic adjustment, with the generated traffic deviating only by about 5% from the real target traffic. This performance significantly exceeds that of current SUMO tools like RouteSampler, which struggle to accurately follow the total daily traffic curve, especially during peak hours when severe traffic congestion occurs.
Download

Paper Nr:	98
Title:	Autonomous Vehicle for Industry 5.0: Digital Twin for System Safety Validation
Authors:	Raivo Sell, Mohsen Malayjerdi, Ehsan Malayjerdi, Mauro Bellone and Heiko Pikner
Abstract:	Autonomy and digitalization are megatrends in today’s world and influence our everyday lives on many levels. The same applies to industry, whereas the manufacturing and engineering industry is heavily under digitalization, also known as Industry 4.0. Now the next step is to focus on where human-centric and sustainable resilient processes are considered a priority. This is an Industry 5.0 paradigm, where humans and robots must work together, with social aspects and increased safety in mind. From the product development and engineering point of view, realistic system simulations and digital counterparts are beneficial to ensure proper complex system development and interactions between robots and humans. In this research, we investigate the methodology to design and implement a comprehensive digital twin of an Autonomous Vehicle (AV) interacting in the context of Industry 5.0 and modern industrial environments. We propose a step-by-step digital twin creation methodology for industrial environments where the AV shuttle bus is intended to serve as a mobility service for the workforce connected to industrial processes. In this research, the main focus is a safety assessment and simulations of an AV interaction in the environment and humans. However, the digital twin, once created, can be used for many other simulations and different purposes.
Download

Paper Nr:	86
Title:	Proposal for Thermal Management Systems for e-Bike Controllers
Authors:	Adriano Figueiredo, José Santos, Tiago Silva and Tiago Gândara
Abstract:	The rise in urban e-bike adoption underscores the need for robust thermal management systems to enhance the efficiency and reliability of critical components such as motor controllers. This study explores compact, lightweight, cost-effective cooling systems to maintain optimal thermal conditions under diverse scenarios. A comprehensive review of cooling technologies, including air, liquid, and hybrid systems, highlights the advantages of phase change materials and heat pipe-based solutions, particularly in combination with forced convection. Control systems leveraging fuzzy logic have emerged as the best solution due to their adaptability and low computational requirements. This research establishes a foundation for integrating innovative thermal management architectures into next-generation e-bikes, promoting sustainability and improved user safety.
Download