SUMO2019:Papers with Abstracts

Abstract. Testing motion planning algorithms for automated vehicles in realistic simulation environments accelerates their development compared to performing real-world test drives only. In this work, we combine the open-source microscopic traffic simulator SUMO with our software framework CommonRoad to test motion planning of automated vehicles. Since SUMO is not originally designed for simulating automated vehicles, we present an inter- face for exchanging the trajectories of vehicles controlled by a motion planner and the trajectories of other traffic participants between SUMO and CommonRoad. Furthermore, we ensure realistic dynamic behavior of other traffic participants by extending the lane changing model in SUMO to implement more realistic lateral dynamics. We demonstrate our SUMO interface with a highway scenario.
Abstract. For emergency vehicle drivers it is an important task to reach the incident location as fast as possible. Therefore a self-organizing green wave could help emergency vehicles to accomplish this goal. This study presents an approach how emergency vehicle can be prioritized at traffic lights and simulates the possible benefit for the emergency vehicle. Traffic data from vehicular communication can be used to find the optimal timing for the traffic light to modify the existing traffic phases and reduce the possible negative impact on other traffic participants.
Abstract. Recent developments such as increasing automation and connectivity of vehicles as well as new regulations for real driving emissions lead to a stronger consideration of traffic and traffic control in automotive development. The increasing complexity of vehicular systems requires a highly virtualized development process. Therefore, a co- simulation solution of DYNA4’s virtual vehicle with SUMO’s microscopic traffic is presented here. Despite increasing automation, virtual test drives often still require a virtual test driver. Thus, the co-simulation solution is extended by combining the driver models of both tools. The operational decision making level of DYNA4 is extended by SUMO’s tactical driver decisions, aiming at virtual test drives in complex surrounding traffic with realistic reaction on traffic and traffic control and reduced parametrization effort. By comparing two variants it is shown that a higher reference speed and more aggressive lane change parameters lead to an increase of usage of the left lane and an increase in achieved speeds.
Abstract. Bangkok is notorious for its chronic traffic congestion due to the rapid urbanization and the haphazard city plan. The Sathorn Road network area stands to be one of the most critical areas where gridlocks are a normal occurrence during rush hours. This stems from the high volume of demand imposed by the dense geographical placement of 3 big educational institutions and the insufficient link capacity with strict routes. Current solutions place heavy reliance on human traffic control expertises to prevent and disentangle gridlocks by consecutively releasing each queue length spillback through inter-junction coordination. A calibrated dataset of the Sathorn Road network area in a microscopic road traffic simulation package SUMO (Simulation of Urban MObility) is provided in the work of Chula-Sathorn SUMO Simulator (Chula-SSS). In this paper, we aim to utilize the Chula-SSS dataset with extended vehicle flows and gridlocks in order to further optimize the present traffic signal control policies with reinforcement learning approaches by an artificial agent. Reinforcement learning has been successful in a variety of domains over the past few years. While a number of researches exist on using reinforcement learning with adaptive traffic light control, existing studies often lack pragmatic considerations concerning application to the physical world especially for the traffic system infrastructure in developing countries, which suffer from constraints imposed from economic factors. The resultant limitation of the agent’s partial observability of the whole network state at any specific time is imperative and cannot be overlooked. With such partial observability constraints, this paper has reported an investigation on applying the Ape-X Deep Q-Network agent at the critical junction in the morning rush hours from 6 AM to 9 AM with practically occasional presence of gridlocks. The obtainable results have shown a potential value of the agent’s ability to learn despite physical limitations in the traffic light control at the considered intersection within the Sathorn gridlock area. This suggests a possibility of further investigations on agent applicability in trying to mitigate complex interconnected gridlocks in the future.
Abstract. In many cases, driving simulator studies target how test persons interact with surround- ing traffic and with traffic signals. Traffic simulations like SUMO specialize in modeling traffic flow, which includes signal control. Consequently, driving and traffic simulation are coupled to benefit from the advantages of both. This means that all except the driven (ego) vehicle are controlled by the traffic simulation. Essential vehicle dynamics data are exchanged and applied frequently to make the test person interact with SUMO-generated traffic. Additionally, traffic lights are controlled by SUMO and transferred to the driving simulation. The system is used to evaluate an Adaptive Cruise Control (ACC) system, which considers current and future traffic light states. Measures include objective terms like traffic flow as well as the subjective judgement of the signal program, the ACC and the simulation environment.
Abstract. In this paper, the focus is put on the integration of XVR, SE-Star and SUMO simulators via the Driver+ test-bed, where XVR provides different learning environments for all levels of incident command, SE-Star handles crowd simulation and SUMO focuses on vehicular simulation and routing. With the test-bed and the provided services these simulation tools can synchronically exchange information with each other, creating a common simulation space that offers more possibilities for CM-training, trials and tests. A simulation scenario around the train station in Rotterdam, the Netherlands, is established for demonstration of the connected systems.
Abstract. Nature and human-made hazards, like hurricanes, inundations, terroristic attacks or in- cidents in nuclear power plants, make it necessary to evacuate large urban areas in a short time. So far, the consideration of railway transportation is rarely part of the evacuation strategies. One of the reasons is the unknown capacity of this infrastructure.
In the case of hurricanes Katrina and Rita (USA) the evacuation was accomplished with pri- vate vehicles and buses. In Germany, especially in the conurbation of Nordrhein-Westfalen, where many roads are overloaded during the daily rush hours, it will not be possible to use only road dependent vehicles like private cars or busses to evacuate a large number of people into save areas.
After the nuclear power plant disaster of Fukushima, the working group ‘AG Fukushima’ was founded, which recommends the use of trains for large-scale emergency evacuations. However, it is not clear if the capacity of train stations is enough to handle these large evacuations in time. Hence, this work deals with the question of how the capacity of train stations can be quantified and optimised for this application. In order to estimate the capacity of train stations we use and further develop the Ju ̈lich Pedestrian Simula- tor (JuPedSim), a software for pedestrian dynamics simulations. Therefore, a model of a train station is built in JuPedSim and several parameters like the inflow and outflow of the pedestrians are examined, to find the best routing strategy and organisational ac- tions inside the station. The focus of this contribution lies in the identification of critical bottlenecks. An estimation of which parameters are influencing congestion at these bottle- necks is presented. Additionally, organisational strategies are outlined, which can prevent congestion and increase the capacity of a train station.
Abstract. Large majority of control methodologies used in traffic applications require short-time prediction of the environment. For instance, in widely-used Model Predictive Control [1] employed to reduce fuel and energy consumption of vehicles in a platoon, information about future velocity profiles of leading vehicles is necessary. In such case, the dynamic model should provide information more detailed than prediction of averaged and global quantities. Additionally, if the control input is to be applied at high-frequencies, traffic model must be solved in a short period of time.
We propose a novel framework which addresses aforementioned problems by estimating the vehicle velocity at any location in the domain based on the real-time information from induction loops downstream. Additionally, our formulation is linear and low-dimensional (i.e. consists of few degrees of freedom) meaning that the estimation can be executed at high frequencies. First a mapping is constructed from velocities at discrete locations to the smooth continuous field, which is subsequently projected onto its most significant principal components. Next, current state of such system is estimated using Kalman filter by combining the linear, wave-like dynamics of the traffic with the instantaneous information provided by induction loops. Short-term traffic prediction is then achieved by integration of the model forward in time.
The proxy methodology is validated using SUMO simulation on the test case of the vehicles approaching a traffic junction. The performance is evaluated based on sampling reconstructed continuous waveform at the locations and timestamps of the vehicles in the reference data and calculating velocity errors. Separate cases are considered where drivers follow Intelligent Driver Model perfectly and with varying levels of uncertainty.
Abstract. The recent advances in adaptive control and autonomous vehicles have given rise to the studies on cooperative control of road vehicles, and the consequent effects on traffic flow performances. In this paper, we summarize our findings from a simulation-based solution of a problem that seeks the joint optimization of a number of link-based performances of vehicular traffic flow considering explicitly the emissions exhausted using the Eclipse SUMO micro-simulation environment in order to discuss the effectiveness of the penetration rates of cooperatively controlled vehicles in mixed traffic.
Abstract. The center of the Kadıköy area in Istanbul is extremely crowded due to overlap of the terminals for the subway and the marine transit lines. When the Kadıköy-Kartal subway’s terminal is added in the middle of the Kadıköy in 2013 without the analysis on crowd dynamics, vehicular and pedestrian traffic have become much more complicated to be efficiently managed. When the pedestrians have a crossing gap, most of them make the decision of crossing without considering the signal phase. Likewise, when it is a pedestrian clearance phase, there can be situations where all the pedestrians cannot cross the street because of high density and insufficient green time. We therefore propose an adaptive traffic control system considering the traffic flows of road vehicles and pedestrians with field data. We have utilized the Eclipse SUMO micro-simulator in conjunction with TraCI for modeling the case. Comparison of fixed time and adaptive signal controllers is provided. Simulations have shown that, reductions in the delay time for both vehicles and pedestrians are achieved by using adaptive signal controllers.
Abstract. Starting from the problems of nowadays’ urban traffic (congestions, imperfect timing of traffic lights, high impact of lane changes) we investigate the feasibility of a cooperative intelligent agent based solution as an overall control scheme governing the car flow in congested urban intersections.
The proposed complex solution features both the intelligent traffic control and the car platooning. In order to test and verify the merits of the proposed solution in urban intersection of a widely variable topology, but also to support our future research aims, a simulation platform, extending the basic functionalities of SUMO with the options of intelligent communication and cooperative co-acting, was designed and developed.
Abstract. Transitions of Control (ToC) play an important role in the simulative impact assessment of automated driving because they may represent major perturbations of smooth and safe traffic operation. The drivers' efforts to take back control from the automation are accompanied by a change of driving behavior and may lead to increased error rates, altered headways, safety critical situations, and, in the case of a failing takeover, even to minimum risk maneuvers. In this work we present modeling approaches for these processes, which have been introduced into SUMO recently in the framework of the TransAID project. Further, we discuss the results of an evaluation of some hierarchical traffic management (TM) procedures devised to ameliorate related disturbances in transition areas, i.e., zones of increased probability for the automation to request a ToC.
Abstract. The electrification of transport is one of the key parts of the present aim to reduce undesirable vehicular emissions in the atmosphere. While the full electrification of personal vehicles is mostly associated with employing a big battery pack on the board and charging on (static) charging stations, another interesting possibility appears in the case of public transport – dynamic drawing of the power from overhead wires. Regarding vehicles moving on the road, this concept is used by trolleybuses or hybrid trolleybuses, i.e. vehicles combining power from the overhead wires and batteries.
A replacement of classic buses (with a combustion engine) with (hybrid) trolleybuses is hardly possible without an appropriate adjustment of public transport lines and the necessary infrastructure. For this purpose, a simulation of the adjusted public transport service may be used to identify weaknesses of the proposed solution.
This paper presents a new vehicle device and a new additional part of road infrastructure in SUMO. It introduces device.elecHybrid based on existing device.battery, extending its functionality and tailoring it for the needs of hybrid trolleybuses. In addition, overhead wires and traction substations are implemented. As the voltage and electric cur- rents in the overhead wires depend on traffic, the overhead wire parameters are optionally evaluated by a built-in electric circuit solver using Kirchhoff’s laws.
The proposed changes allow us to simulate hybrid trolleybus in-motion charging under the overhead wire. The extensions can be immediately used in micro-simulations or even (in a simplified version) in the meso-simulation mode.
Abstract. Traffic congestion on not only highways but also complex urban road networks has attracted the attention of many researchers. Traffic congestion growing in urban road net- works is an inevitably important problem especially for populated cities during rush hours. A traffic blockage can be realized as the source of traffic congestion, which can propagate to form queues and sometimes a gridlock. Traffic blockages are triggered by complicated factors ranging from temporal and spatial situations. Recurrent congestion is a traffic congestion that occurs during morning and evening rush hours e.g. from school buses and parent vehicles to drive their children to-and-from schools. In addition, unforeseen, unexpected events that can cause as non-recurrent traffic congestion e.g. car breakdowns, accidents, road maintenance, and severe weather conditions, which can disorder normal traffic flows and reduce road capacity. Traffic blockage may spread its negative impacts to neighbouring upstream and downstream links. And that can lead to the formation of congestion gridlock, which further reduce traffic flow efficiency in a complex urban road network. These problems are vital but often tough to resolve in urban road networks. In this paper, the Chula-Sathorn SUMO Simulator (Chula-SSS) dataset has been used with Simulation of Urban Mobility (SUMO) to simulate recurrent and non-recurrent congestion cases. The detection is based on the information from simulated lane area detectors. For non-recurrent case, lanes are closed to simulate the gridlock occurrences. With the morning case of calibrated Chula-SSS dataset, both recurrent and nonrecurrent congestion based gridlock have been studied with upstream and downstream nearby detectors and preliminary results are herein reported upon the gridlock status as detected by using different combinations of traffic jam length and mean speed conditions at both the upstream and downstream detectors of every intersection within the critical looped road segments.
Abstract. The introduction of highly automated driving functions is one of the main research and development efforts in the automotive industry worldwide. In the early stages of the development process, suppliers and manufacturers often wonder whether and to what extend the potential of the systems under development can be estimated in a cheap and timely manner. In the context of a current research project, a sensor system for the detection of the road surface condition is to be developed and it is to be investigated how such a system can be used to improve higher level driving functions. This paper presents how road surface conditions are introduced in various elements of the microscopic traffic simulation such as the actual network, the network editor, a device for detection, and an adaptation of the standard Krauß car following model. It is also shown how the adaptations can subsequently affect traffic scenarios. Furthermore, a summary is given how this preliminary work integrates into the larger scope of using SUMO as a tool in the process of analyzing the effectiveness of a road surface condition sensor.
Abstract. Bicycle traffic is becoming an increasingly important part of urban traffic. Thus, the simulation and accurate representation of bicycle traffic in microscopic traffic simulation software is gaining importance. As bicycle traffic increases, dedicated bicycle infrastructure is designed to accommodate bicycle traffic. Especially at intersections, the design of intersection approaches follows specific rules and geometric limitations as defined by official design guidelines used in different countries across the world. However, when special environmental factors that affect the intersection layout, such as available space or gradient are not considered, specific standard forms of intersection approaches can be determined based on the number of traffic lanes, the traffic signal control and in the case of this study, the availability as well as the type of dedicated bicycle infrastructure. Categories with available bicycle infrastructure include the cases of bicycle lanes or advisory cycle lanes with advance stop lines for direct left turning bicyclists, the bicycle lanes or advisory bicycle lanes with bicycle boxes and bicycle lanes or bicycle paths with advanced stop lines and a stop area downstream for facilitating an indirect left turn or a two-stage (left) turn of bicyclists. The simulation of such bicycle infrastructure is not natively supported in microscopic traffic simulation software and is mostly only possible through intuitive adjustment of existing network design elements. In this paper, fictional intersections with special bicycle infrastructure are modelled in SUMO. Bicycle traffic data is collected at intersections in Germany with different types of bicycle infrastructure. The collected bicycle traffic data is then used to evaluate the intersection models. Specific recommendations for modelling bicycle infrastructure at intersection approaches in SUMO are provided, and limitations of the proposed methodologies and software limitations are discussed. Results show that the developed solutions can be used to model the bicycle traffic behavior with a reasonable degree of accuracy only for simulation scenarios and traffic situations unaffected by the identified software limitations.
Abstract. Operational behavior models are used in traffic simulations to represent the subconscious, short-termdecisions made by road users to respond to other road users, the infrastructure and traffic control measures. Calibration and validation of these models can be achieved using observed trajectory data from real road users. For lane bound traffic, it is assumed that road users intend to follow a given lane with a certain desired speed across the intersection. Any deviation from this planned path is in response to other road users or the environment. It is difficult, however, to identify and separate the desired movement of more flexible road users that do not follow lane disciple, such as bicyclists, from movements made as a reaction to other road users or obstacles. This can lead to poor calibration of operational behavior models and unrealistic behavior in the simulation. Tactical behavior models recreate the conscious decisions made on a time horizon of seconds to minutes to cope with the immediate traffic situation. As such, tactical behavior models are responsible for selecting the planned path across an intersection.
Here, SUMO is coupled with the simulation software DYNA4 to create a simulated road environment for a bicycle simulator. Trajectories observed in reality are displayed as potential prescribed pathways across the simulated intersection. Participants in the simulator study are instructed to select and follow one of the prescribed pathways as closely as possible while responding naturally to other road users and obstacles in the environment. The resulting trajectory data is used to calibrate existing operation al and tactical path finding behavior models for bicyclists at signalized intersection.
Abstract. The project MAVEN (see, funded by the European Com- mission, aims at developing a system for infrastructure-assisted platoon organization and green phase negotiation for automated connected vehicles (ACVs). Vehicle-to-Everything (V2X) communication protocols are hereby used for the insertion of vehicles into a traffic simulation of a real-world intersection. Until now, real world traffic could be inserted into a simulation through stationary detectors, for example magnet field sensors, induction loops, cameras, radar etc. The downside of this detection method is that only momentary information can be obtained and e.g. the behavior of the vehicles approaching an intersection can only be approximated. ACVs however continuously broadcast their positions and speeds via CAMs. Detecting vehicles though these messages leads to a more realistic representation of the vehicle’s driving behavior. The current paper describes how CAMs are used to place and move ACVs inside the simulation of a real-world intersection in Braunschweig with the traffic simulation SUMO (Simulation of Urban Mobility). Furthermore, it describes an approach to how these continuously detected vehicles could be further used as control units. Since the positions and speeds of ACVs are synchronized with the real-world behavior, they can be used to adjust the simulated upstream movements and positioning of conventional vehicles (CV) to match reality. Until all vehicles are equipped with V2X technology, this approach could enable more realistic simulated traffic flow behavior.
Abstract. The emerging usage of connected vehicles promises new business models and a high level of innovation, but also poses new challenges for the automotive domain and in particular for the connectivity dimension, i. e. the connection between vehicles and cloud environments including the architecture of such systems. Among other challenges, IoT Cloud platforms and their services have to scale with the number of vehicles on the road to provide functionality in a reliable way, especially when dealing with safety-related functions. Testing the scalability, functionality, and availability of IoT Cloud platform architectures for connected vehicles requires data from real world scenarios instead of hypothetical data sets to ensure both the proper functionality of distinct connected vehicle services and that the architecture scales with a varying number of vehicles. However, the closed and proprietary nature of current connected vehicle solutions aggravate the availability of both vehicle data and test environments to evaluate different architectures and cloud solutions. Thus, this paper introduces an approach for connecting the Eclipse SUMO traffic simulation with the open source connected vehicle ecosystem Eclipse Kuksa. More precisely, Eclipse SUMO is used to simulate traffic scenarios including microscopic properties like the position or emission. The generated data of each vehicle is then be sent to the message gateway of the Kuksa IoT Cloud platform and delegated to an according example service that consumes the data. In this way, not only the scalability of connected vehicle IoT architectures can be tested based on real world scenarios, but also the functionality of cloud services can be ensured by providing context-specific automotive data that goes beyond rudimentary or fake data-sets.
Abstract. Expectations are that automated and connected mobility will increase road safety and traffic efficiency. However, due to possible shortcomings of new technologies , road users may be confronted with disturbances and potential safety risks. The mitigation of such risks will bring necessary changes to road infrastructure, vehicles and road-users’ behavior. In a traffic environment that was built to fit the human perception, preemptive simulation of parametrized scenarios can provide guidelines for what changes and difficulties are to be expected. Utilizing SUMO in varied scenarios, this paper outlines the creation of virtual models that correspond to interaction hot spots on the Austrian road network - from digitizing the infrastructure, to calibrating a simulation scenario with congruent traffic measurements - while it concludes with the evaluation of scenario simulation results. The approach is demonstrated for a selected motorway ramp scenario, varying rates of automated vehicles and different infrastructure layouts. Performance indicators like vehicle speed distributions and traffic disruptions are defined and analyzed to investigate how adaptations can mitigate risks, influence traffic flow and hence support progressing vehicle automation.
Abstract. The co-ordination between traffic signals is assumed to be important for the good organization of a transport system. By using an artificial approach to create and analyze a multitude of transportation systems, a few different simple traffic signals programs has been put to the test and compared to each other. The result is that a well co-ordinated system can be outperformed by a non-coordinated signal set-up, where all signals controlers run in (single intersection) actuated mode. Clearly, these results are preliminary and require more investigation.
Abstract. In the last decade, many efforts to solve traffic congestion and sustainable growth issues are going in the direction of research and investments in smart cities and consequently smart mobility.
We use the proposed simulation framework is compatible with SUMO 1.1.0. We use it to study multi-modal commuting and parking optimization issues in a state-of-the-art large-scale mobility scenario, and we intend to demonstrate the ease of use and its capabilities.