This work gives a short impression on the long history of SUMO's default car-following model, the one according to Stefan Krauß. It tries to summarize some of the ideas that were needed to change a beautiful theoretical model into a versatile and reliable tool that does its job in a wide variety of different situations, almost without producing obnoxious surprises.
Trajectory data are great data to work with, since they are the most natural data for traffic. However, they provide considerable challenges when tried to put into a micro-simulation framework such as SUMO. This work here gives an example what had to be done to arrive at a simulation that is driven by these data. Succeeding in this, microscopic tools can be much better tested against real data.
Urban traffic management is a critical challenge in modern cities, necessitating innovative solutions to optimize traffic flow and reduce congestion. This research presents the development of an AI engine leveraging spatio-temporal learning techniques for urban traffic calibration. The proposed methodology leverages digital twin scenarios driven by microscopic simulations, which capture detailed vehicle behaviors—including interactions, lane changes, and driver dynamics to provide granular insights into urban traffic patterns. At the core of the AI engine is the Dynamic Spatio-Temporal Graph Attention Network (DSTGAT), a hybrid model that combines multi-head Graph Attention Networks (GATv2) with Long Short-Term Memory (LSTM) networks. DSTGAT exploits the joint spatio-temporal relationships inherent in traffic data by processing sequential snapshots of urban traffic, where each snapshot is represented as a graph with nodes indicating urban zones and edges carrying continuous flow values. The GATv2 layers, enhanced with residual connections and batch normalization, extract robust spatial embeddings, while the LSTM aggregates these embeddings over time to capture dynamic patterns and predict future traffic flows in real-time. The AI engine incorporates an iterative feedback loop that continuously refines the OD demand using synthetic scenarios, improving estimation accuracy across diverse urban environments. Preliminary results show that the DSTGAT‑based framework lowers OD‑estimation error on simulated data, suggesting its usefulness as an input to downstream traffic‑management strategies.
The growing number of battery electric vehicles (BEV) implies changes in urban infrastructure. Large amounts of charging stations are to be built within few years to supply energy for a mostly electric vehicle fleet. In parallel the power grid has to be adpated to the growing energy demand. However charging behaviour depends on mobility patterns and which type of charging stations can be accessed. For example, charging at work is mostly restricted to employees and happens during working hours. Public charging stations cater any BEV user but take precious public space. In this work several charging station configurations for a city are studied on a microscopic level together with different BEV shares and charging strategies and policies. 24 hour working day SUMO microsopic traffic demand travels along trip chains and charges their electric vehicle battery when needed. Parking and charging infrastructure is limited to reasonable capacity for a medium sized German city. Then the configurations are evaluated with respect to charging station usage and the expected energy demand/supply of BEV for balanced and time-shift charging strategies as well as vehicle-to-home/vehicle-to-grid policies. Applying these strategies and policies on a large scale can lower the energy peak demand visibly.
Traffic simulations play a crucial role in urban planning and mobility management by providing insights into transportation systems. However, their effectiveness heavily depends on accurate demand calibration, often requiring large amounts of observational data. This poses a challenge in settings with limited data availability. In this paper, we propose a methodology for calibrating SUMO scenarios under data-scarce conditions. To contextualize our approach, we first review existing SUMO scenarios and their demand calibration strategies. We then introduce the Mannheim SUMO Traffic Model (MaST) as a case study and employ the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) to optimize route probabilities as input for the existing routeSampler tool provided by SUMO. Results indicate that our method significantly improves calibration accuracy compared to baseline approaches both for 3-hour and 24-hour scenarios. While our findings suggest that the proposed methodology can support demand calibration in data-limited environments, further research is needed to assess its generalizability and effectiveness in different contexts.
Microscopic traffic simulation tools enable the simulation of various traffic demands and their impacts. By implementing realistic traffic models, they are commonly used to analyze traffic engineering measures. Since microscopic traffic simulations focus more on the network than on individual small-scale scenarios, such frameworks are appropriate for efficiency studies. In this respect, SUMO allows only a limited investigation of efficiency measures at the vehicle level, as the underlying vehicle models, such as Krauss, Gipps, or IDM, do not regard vehicle dynamics constraints. To encounter this disadvantage, SUMO is enriched by the elements of the submicroscopic simulation tool ASM from dSPACE, including a highly detailed vehicle dynamics model for the ego vehicle. Combining the elements of microscopic and submicroscopic simulation in a co-simulation framework, the effectiveness assessment of efficiency measures can be broken down to vehicle level. Besides the efficiency aspect, the proposed co-simulation bridge also offers the possibility to simulate sensing technology in a 3D environment with AURELION, where SUMO provides realistic traffic situations.
Short dwell times at bus stops are crucial for efficient public transport operations, yet existing traffic simulation tools commonly simplify passenger boarding. In this paper, we extend the SUMO-based co-simulation framework, Sumonity, to incorporate a sub-microscopic pedestrian model for city bus boarding. Our approach simulates real-time passenger flow, pathfinding, and door congestion in a Unity-based environment. We conduct a full-factorial simulation experiment with four bus door configurations, ranging from fully open double doors to partially closed options. We also consider different passenger loads between 1 and 50, yielding 200 unique scenarios. Detailed spatiotemporal data on passenger movements and boarding times are generated for each scenario. Analysis of crowding behaviors and door usage reveals significant sensitivity of boarding times to both passenger volume and door availability. These findings demonstrate the importance of accurately modeling pedestrian interactions for reliable dwell-time forecasts and underscore the potential of sub-microscopic pedestrian simulations.
Realistic and comprehensive traffic simulations are essential for the effective testing and evaluation of emerging technologies, such as Vehicle-to-X (V2X) communication, and diverse use cases, particularly within complex urban environments. While current traffic scenarios often focus on motorized vehicles, there is a need to address the safety of vulnerable road users (VRUs), such as pedestrians and cyclists. This is especially relevant in light of the European Union's Vision Zero initiative, which aims for zero road fatalities by 2050. Although a few pedestrian-focused scenarios exist, there is no scenario specifically addressing bicycle traffic, despite their status as one of the most at-risk VRUs, with stagnant fatality rates in recent years. To address this gap, this paper introduces the Hanover Traffic Scenario for SUMO (HaTS), a novel traffic scenario including motorized vehicles and bicycles. HaTS provides a detailed and accurate representation of the road network, traffic light systems, and buildings within the city center of Hanover, Germany. A key feature of HaTS is its integration of real-world traffic count data for both bicycles and motorized vehicles, enabling a realistic and representative traffic demand representation. Additionally, a novel metric is employed for the parametrization of the scenario, enhancing the alignment between real and simulated traffic volumes. For the validation we compare the results of the HaTS with the real world traffic counts. HaTS is the first open-source SUMO scenario focused on bicycles, providing a realistic representation of the road network and traffic demand, thereby contributing to the advancement of urban traffic simulations.
The digital transformation of the transport sector in our cities will be led by the deployment of large-scale digital twins, interacting with their real world counterpart to model, predict, and improve movements and reoccurring patterns. Traffic simulation is an essential tool in this area. While both macroscopic and microscopic simulations are possible, only the latter provide enough detail to realize sophisticated Intelligent Traffic Systems (ITS). One of the biggest challenges is accurately modeling road traffic on a large scale due to limitations in both reliable data sources, as well as the quickly increasing complexity of size. Only a handful of city-scale traffic scenarios exist, and only a few of them include public transport modalities. With this paper, our aim is to extend this list by integrating bus traffic within the Berlin SUMO Traffic scenario (BeST). We provide an overview of potential data sources and a detailed description of the applied methodology. As the scenario was initially calibrated with only individual private traffic, we conduct an evaluation on how the added traffic volume affects the stability of the scenario.
In recent years, Unmanned Aerial Vehicles (UAVs) have emerged as effective tools for traffic monitoring and control by offering high-resolution, aerial observations of vehicular movement. Although UAV simulation is well established, tools to capture microscopic traffic measurements from UAV-based observations remain limited. This paper introduces SUMO-UAV-Py, an open-source SUMO plugin that integrates UAV-based sensing into microscopic traffic simulations in Python. SUMO-UAV-Py captures detailed vehicle observations by dynamically employing multiple UAVs to observe traffic measurements based on their position and field-of-view (FoV). Performance evaluations on a mid-sized network demonstrate that SUMO-UAV-Py maintains simulation performance comparable to standard post-processing methods, confirming its suitability for large-scale traffic monitoring research.
Swift advances in computing and artificial intelligence (AI) technologies of late have prompted the increasing applications of digital twins (DiTs) to various sectors for boosting effectiveness and productivity. DiTs have been envisioned to possess immense potential for transforming numerous domains and sectors in the recent report of National Academics due to their powerful real-time decision-making based on modeling and simulating physical systems. This paper deals with a novel design of digital twin-aided municipal traffic control (DiTAT) for best traffic management over a targeted municipal region, based on real-world traffic video imagery gathered by available roadside surveillance cameras. DiTAT analyzes sequences of video frames to extract traffic volume details, including the start time, speed, incoming zone, and outgoing zone of every vehicle in existence. Being DiT-based, DiTAT employs the automated, open-source traffic simulator (SUMO) as the digital twin of the physical roadway configuration over the target region to try various traffic light control settings under the extracted traffic volume details for identifying the most favorable setting. The identified setting is then sent to the physical roadway traffic lights for realization to manage traffic during the next time window, when its resulting traffic is simulated by SUMO again to get the best setting for the subsequent time window reactively. This process repeats continuously window by window, with a bidirectional interplay between SUMO simulation and physical traffic for the target region. DiTAT is demonstrated to lift transport performance under real-world traffic scenarios.
Mobility in metropolitan regions is changing. The distribution of space in cities, the design of transport modes, and the organization of mobility are being re-thought. However, no matter the changes and innovations on the way to a more sustainable future, essential constants must be upheld: In the event of minor, regionally limited emergencies, medical assistance must reach those in need quickly. When dealing with large-scale emergencies, the ability to evacuate the area promptly must be ensured. The impact analysis of mobility innovations on emergency services within urban areas so far has been based purely on empirical observations using existing data. Currently, it is only possible to analyze what-if considerations in a limited way. Nevertheless, due to the increasingly rapid changes in mobility, a comprehensive and interlinked analysis will be necessary. This is the key contribution of rescuePY: rescuePY is a simulation suite based on the mesoscopic and microscopic simulation environment hybridPY. It allows holistic and microscopic transport modeling of rescue infrastructure to quantify the impact of the mobility transition towards higher sustainability on the performance of rescue services. The main features of this software are: Rescue system assessment for strategic, long-term planning Mobility-influence studies for operative, mid-term planning Activity-based urban evacuation modeling The capabilities of rescuePY are demonstrated by two applications: a simulation- based, mesoscopic system analysis of emergency services in Munich compared to real-world data and microscopic modeling of emergency vehicles (EMVs) in different road architectures. Ongoing developments aim to improve the evaluation methodology for the aggregated impact analysis of mobility innovations on rescue response services.
The UNECE regulation R157 [1] describes the requirements for a successful implementation of an approvable ALKS (Automated Lane-Keeping System) in great detail. This paper reviews some of the content of this document and describes the first steps that would be needed on how to implement such an ALKS as another driver model into the open source microscopic traffic flow simulator SUMO.
