Behzad Bamdad Mehrabani, L. Sgambi, S. Maerivoet, M. Snelder
Development of large-scale traffic simulation models have always been challenging for transportation researchers. One of the essential steps in developing traffic simulation models, which needs lots of resources, is travel demand modeling. Therefore, proposing travel demand models that require less data than classical travel demand models is highly important, especially in large-scale networks. This paper first presents a travel demand model named as probabilistic travel demand model, then it reports the process of development, calibration and validation of Belgium traffic simulation model. The probabilistic travel demand model takes cities' population, distances between the cities, yearly vehicle-kilometer traveled, and yearly truck trips as inputs. The extracted origin-destination matrices are imported into the SUMO traffic simulator. Mesoscopic traffic simulation and the dynamic user equilibrium traffic assignment are used to build the base case model. This base case model is calibrated using the traffic count data. Al-so, the validation of the model is performed by comparing the real (extracted from Google Map API) and simulated travel times between the cities. The validation results ensure that the model is a superior representation of reality with a high level of accuracy. The model will be helpful for road authorities, planners, and decision-makers to test different scenarios, such as the im-pact of abnormal conditions or the impact of connected and autonomous vehicles on the Belgium road network.
{"title":"Development, Calibration, and Validation of a Large-Scale Traffic Simulation Model: Belgium Road Network","authors":"Behzad Bamdad Mehrabani, L. Sgambi, S. Maerivoet, M. Snelder","doi":"10.52825/scp.v4i.199","DOIUrl":"https://doi.org/10.52825/scp.v4i.199","url":null,"abstract":"Development of large-scale traffic simulation models have always been challenging for transportation researchers. One of the essential steps in developing traffic simulation models, which needs lots of resources, is travel demand modeling. Therefore, proposing travel demand models that require less data than classical travel demand models is highly important, especially in large-scale networks. This paper first presents a travel demand model named as probabilistic travel demand model, then it reports the process of development, calibration and validation of Belgium traffic simulation model. The probabilistic travel demand model takes cities' population, distances between the cities, yearly vehicle-kilometer traveled, and yearly truck trips as inputs. The extracted origin-destination matrices are imported into the SUMO traffic simulator. Mesoscopic traffic simulation and the dynamic user equilibrium traffic assignment are used to build the base case model. This base case model is calibrated using the traffic count data. Al-so, the validation of the model is performed by comparing the real (extracted from Google Map API) and simulated travel times between the cities. The validation results ensure that the model is a superior representation of reality with a high level of accuracy. The model will be helpful for road authorities, planners, and decision-makers to test different scenarios, such as the im-pact of abnormal conditions or the impact of connected and autonomous vehicles on the Belgium road network.","PeriodicalId":439794,"journal":{"name":"SUMO Conference Proceedings","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128705204","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Aboozar Roosta, Heather Kaths, Mirko Barthauer, J. Erdmann, Yun-Pang Flötteröd, M. Behrisch
Microscopic traffic simulation tools provide ever-increasing value in the design and implementation of motor vehicle transport systems. Research and development of automated and intelligent technologies have highlighted the usefulness of simulation tools and development efforts have accelerated in recent years. However, the majority of traffic simulation software is developed with a focus on motor vehicle traffic and has limited capabilities in the simulation of bicycles and other micro-mobility modes. Bicycles, e-bikes and cargo bikes represent a non-negligible modal share in many urban areas and their impact on the operation, efficiency and safety of traffic systems must be considered in any comprehensive study. The Differentiation between different types of micro-mobility modes, including microcars, e-kick scooters, different types of bicycles and other personal mobility devices, has not yet attracted enough attention in the development of simulation software which creates difficulties in including these modes in simulation-based studies. On November 25th, 2022, members of the SUMO team at DLR organized a workshop to assess the state of bicycle simulation in SUMO, identify shortcomings and missing capabilities and prioritize the order in which bicycle traffic related features should be modified or implemented in the future. In this paper, different aspects of simulating bicycle traffic in SUMO are examined and an overview of the results of the workshop discussions is given. Some suggestions for the future development of SUMO emerging from this workshop, are presented as a conclusion.
{"title":"State of Bicycle Modeling in SUMO","authors":"Aboozar Roosta, Heather Kaths, Mirko Barthauer, J. Erdmann, Yun-Pang Flötteröd, M. Behrisch","doi":"10.52825/scp.v4i.215","DOIUrl":"https://doi.org/10.52825/scp.v4i.215","url":null,"abstract":"Microscopic traffic simulation tools provide ever-increasing value in the design and implementation of motor vehicle transport systems. Research and development of automated and intelligent technologies have highlighted the usefulness of simulation tools and development efforts have accelerated in recent years. However, the majority of traffic simulation software is developed with a focus on motor vehicle traffic and has limited capabilities in the simulation of bicycles and other micro-mobility modes. Bicycles, e-bikes and cargo bikes represent a non-negligible modal share in many urban areas and their impact on the operation, efficiency and safety of traffic systems must be considered in any comprehensive study. The Differentiation between different types of micro-mobility modes, including microcars, e-kick scooters, different types of bicycles and other personal mobility devices, has not yet attracted enough attention in the development of simulation software which creates difficulties in including these modes in simulation-based studies. On November 25th, 2022, members of the SUMO team at DLR organized a workshop to assess the state of bicycle simulation in SUMO, identify shortcomings and missing capabilities and prioritize the order in which bicycle traffic related features should be modified or implemented in the future. In this paper, different aspects of simulating bicycle traffic in SUMO are examined and an overview of the results of the workshop discussions is given. Some suggestions for the future development of SUMO emerging from this workshop, are presented as a conclusion.","PeriodicalId":439794,"journal":{"name":"SUMO Conference Proceedings","volume":"435 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126985221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Developing new solutions to complicated large-scale problems typically requires large-scale numerical simulation. Therefore, traffic simulations often run against randomized simulations instead of real-world traffic situations. This paper demonstrates a method to calculate the statistical significance of numerical simulations and optimizations in the presence of numerous random variables in complex systems using one-sided paired t-tests. While the paper covers a specific Fujitsu traffic-optimization project which uses SUMO for simulating the traffic situation, the method can be applied to many similar projects where a complete investigation of the solution space is not feasible due to the size of the solution space.
{"title":"Coping with Randomness in Highly Complex Sys-tems Using the Example of Quantum-Inspired Traffic Flow Optimization","authors":"Maria Haberland, L. Hohmuth","doi":"10.52825/scp.v4i.216","DOIUrl":"https://doi.org/10.52825/scp.v4i.216","url":null,"abstract":"Developing new solutions to complicated large-scale problems typically requires large-scale numerical simulation. Therefore, traffic simulations often run against randomized simulations instead of real-world traffic situations. This paper demonstrates a method to calculate the statistical significance of numerical simulations and optimizations in the presence of numerous random variables in complex systems using one-sided paired t-tests. While the paper covers a specific Fujitsu traffic-optimization project which uses SUMO for simulating the traffic situation, the method can be applied to many similar projects where a complete investigation of the solution space is not feasible due to the size of the solution space.","PeriodicalId":439794,"journal":{"name":"SUMO Conference Proceedings","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116235727","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Microscopic traffic flow simulations as tools for enabling detailed insights on traffic efficiency and safety gained numerous popularity among transportation researchers, planners and engineers in the first to decades of the 21st century. By implementing a test bed for simulation scenarios of complex urban transportation infrastructure it is possible to inspect specific effects of introducing small infrastructural changes related to the built environment and to the introduction of advanced traffic control strategies. The possibility of reproducing present problems or the transportation services, such as the ones of public bus services is a key motivation of this work. In this research, we reproduce the road network of the city of Kyoto for observing specific travel patterns of public buses such as the bus bunching phenomena. Therefore, a selection of currently available data sets is used for calibrating a cutout of the Kyoto road network of a relatively large extent. After introducing a method for geodata extraction and conversion, we approach the calibration by introducing virtual detectors representing present inductive loops and make use of historical traffic count records. Additionally, we introduce bus routes partially contributed by volunteer mappers (OSM project). First simulation outcomes show numerous familiar (local knowledge) flow patterns.
{"title":"Generating and Calibrating a Microscopic Traffic Flow Simulation Network of Kyoto","authors":"A. Keler, Wenzhe Sun, Jan-Dirk Schmöcker","doi":"10.52825/scp.v4i.226","DOIUrl":"https://doi.org/10.52825/scp.v4i.226","url":null,"abstract":"Microscopic traffic flow simulations as tools for enabling detailed insights on traffic efficiency and safety gained numerous popularity among transportation researchers, planners and engineers in the first to decades of the 21st century. By implementing a test bed for simulation scenarios of complex urban transportation infrastructure it is possible to inspect specific effects of introducing small infrastructural changes related to the built environment and to the introduction of advanced traffic control strategies. The possibility of reproducing present problems or the transportation services, such as the ones of public bus services is a key motivation of this work. In this research, we reproduce the road network of the city of Kyoto for observing specific travel patterns of public buses such as the bus bunching phenomena. Therefore, a selection of currently available data sets is used for calibrating a cutout of the Kyoto road network of a relatively large extent. After introducing a method for geodata extraction and conversion, we approach the calibration by introducing virtual detectors representing present inductive loops and make use of historical traffic count records. Additionally, we introduce bus routes partially contributed by volunteer mappers (OSM project). First simulation outcomes show numerous familiar (local knowledge) flow patterns.","PeriodicalId":439794,"journal":{"name":"SUMO Conference Proceedings","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114067019","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this study, the physical principles governing car-following (CF) behavior and their impact on traffic flow at signalized intersections are investigated. High temporal-resolution radar data is used to provide valuable insights into actual CF behavior, including acceleration, deceleration, and time headway distribution. Demand-calibrated SUMO simulations are run using empirical CF parameter distributions, and three CF models are evaluated: IDM, EIDM, and Krauss. By emulating radar data in SUMO and processing simulated vehicle traces, discrepancies between empirical and simulated parameter distributions are identified. Further analysis includes comparisons with default SUMO CF model parameters. The findings reveal that measured accelerations differ from CF model parameter accelerations and using the empirical value ($mu = 0.89m/s^2$) leads to unrealistic simulations that fail volume-based calibration. Default parameters for all three models reasonably approximate the mean and median of measured parameters, but fail to capture the true distribution shape, partly due to homogeneity when using default parameters. The results show that it is more effective to simulate with the default parameters provided by SUMO rather than using measurements of real-world distributions without additional calibration. Future work will investigate closing the loop between the measured real-world and SUMO distributions using traditional calibration tactics, as well as assess the impact of calibrated vs. default CF parameters on simulation outputs like fuel consumption.
{"title":"Comparing Measured Driver Behavior Distributions to Results from Car-Following Models using SUMO and Real-World Vehicle Trajectories from Radar","authors":"Max Schrader, Mahdi Al Abdraboh, J. Bittle","doi":"10.52825/scp.v4i.214","DOIUrl":"https://doi.org/10.52825/scp.v4i.214","url":null,"abstract":"In this study, the physical principles governing car-following (CF) behavior and their impact on traffic flow at signalized intersections are investigated. High temporal-resolution radar data is used to provide valuable insights into actual CF behavior, including acceleration, deceleration, and time headway distribution. Demand-calibrated SUMO simulations are run using empirical CF parameter distributions, and three CF models are evaluated: IDM, EIDM, and Krauss. By emulating radar data in SUMO and processing simulated vehicle traces, discrepancies between empirical and simulated parameter distributions are identified. Further analysis includes comparisons with default SUMO CF model parameters. The findings reveal that measured accelerations differ from CF model parameter accelerations and using the empirical value ($mu = 0.89m/s^2$) leads to unrealistic simulations that fail volume-based calibration. Default parameters for all three models reasonably approximate the mean and median of measured parameters, but fail to capture the true distribution shape, partly due to homogeneity when using default parameters. The results show that it is more effective to simulate with the default parameters provided by SUMO rather than using measurements of real-world distributions without additional calibration. Future work will investigate closing the loop between the measured real-world and SUMO distributions using traditional calibration tactics, as well as assess the impact of calibrated vs. default CF parameters on simulation outputs like fuel consumption.","PeriodicalId":439794,"journal":{"name":"SUMO Conference Proceedings","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127681477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Cyclists pose an interesting challenge in the microscopic modeling and simulation of urban traffic. Like motorists, cyclists can move on roadways, tend to have one main axis of movement (longitudinal), and cannot change their velocity instantaneously. However, like pedestrians, cyclists are less bound by lane discipline and are often less rule-oriented than motorists. They flexibly adjust their lateral position within a lane, fluidly move between different types of infrastructure (bicycle lane, sidewalk, roadway), and tactically select their pathways across intersections. Their interactions with other road users are more intuitive and less defined by the lane markings. How should the behavior of such adaptable road users be modeled? In SUMO, modifications to the simulation environment enable the application of car-based models to cyclists. A driving lane is divided into multiple sub-lanes along the longitudinal axis. Lane change and car-following models can be calibrated and applied to simulate realistic bicycle and mixed traffic using this approach. However, the flexible nature of cyclists, particularly at intersections or when switching between different types of infrastructure, is difficult to simulate. A modeling framework for linking the paradigms used to simulate motor vehicle traffic (one-dimensional lane-based models) and pedestrian traffic (two-dimensional social force type models) is presented. Guidelines are used to lead each cyclist through the network while they move freely on a two-dimensional plane, their movement and interactions governed by an adapted social force model. The conceptual framework and an openly available Python package CyclistModel are introduced, and advantages and possible use cases are discussed.
{"title":"Framework for Simulating Cyclists in SUMO","authors":"Heather Kaths, Aboozar Roosta","doi":"10.52825/scp.v4i.219","DOIUrl":"https://doi.org/10.52825/scp.v4i.219","url":null,"abstract":"Cyclists pose an interesting challenge in the microscopic modeling and simulation of urban traffic. Like motorists, cyclists can move on roadways, tend to have one main axis of movement (longitudinal), and cannot change their velocity instantaneously. However, like pedestrians, cyclists are less bound by lane discipline and are often less rule-oriented than motorists. They flexibly adjust their lateral position within a lane, fluidly move between different types of infrastructure (bicycle lane, sidewalk, roadway), and tactically select their pathways across intersections. Their interactions with other road users are more intuitive and less defined by the lane markings. How should the behavior of such adaptable road users be modeled? In SUMO, modifications to the simulation environment enable the application of car-based models to cyclists. A driving lane is divided into multiple sub-lanes along the longitudinal axis. Lane change and car-following models can be calibrated and applied to simulate realistic bicycle and mixed traffic using this approach. However, the flexible nature of cyclists, particularly at intersections or when switching between different types of infrastructure, is difficult to simulate. A modeling framework for linking the paradigms used to simulate motor vehicle traffic (one-dimensional lane-based models) and pedestrian traffic (two-dimensional social force type models) is presented. Guidelines are used to lead each cyclist through the network while they move freely on a two-dimensional plane, their movement and interactions governed by an adapted social force model. The conceptual framework and an openly available Python package CyclistModel are introduced, and advantages and possible use cases are discussed.","PeriodicalId":439794,"journal":{"name":"SUMO Conference Proceedings","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125135443","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Traffic simulation tools are used by city planners and traffic professionals over the years for modelling and analysis of existing and future infrastructural or policy implementations. There are numerous studies on emergency vehicle (EV) prioritization in cities all over the world, but every area is unique and requires the data collection and simulation to be done separately. In this case, the focus area is the Mörfelder Landstraße in Frankfurt am Main, Germany, one of the busiest streets in this city. Thestudy illustrates demand modelling, simulation and evaluation of a traffic improvement strategy for EVs. Vehicular traffic such as passenger cars and trams are simulated microscopically. To perform accurate traffic simulation, input data quality assurance and cleansing of Master Data is required. Therefore, the data is adapted to reproduce the real-world scenario and transformed into the readable format for the simulation model. Vehicular demand is calibrated by traffic count data provided by the Frankfurt Traffic Department. To model road traffic and road network, origin destination matrices using the Gravity Mathematical Model and Open Street Maps are generated, respectively. This process is time-consuming and requires effort. However, this process is critical to get realistic results. In the next step, the road traffic is simulated using SUMO (Simulation of Urban mobility). Finally, EV relevant key performance indicators (KPIs): total trip time and total delay time are derived from simulations. The real-world scenario is compared with five alternative scenarios. The comparison of the KPIs revealed that the real-world scenario results in longer travel times compared to the EV-prioritization scenario. In the least case, the overall travel times for EV has decreased significantly and, as we know, in the case of EVs, even a few seconds saved could prove crucial for a person in need.
多年来，城市规划者和交通专业人员使用交通模拟工具对现有和未来的基础设施或政策实施进行建模和分析。在世界范围内，关于应急车辆优先级的研究有很多，但每个地区都是独特的，需要分别进行数据收集和仿真。在这种情况下，重点区域是德国法兰克福的Mörfelder Landstraße，这是这个城市最繁忙的街道之一。该研究阐述了电动汽车交通改善策略的需求建模、仿真和评估。车辆交通，如乘用车和有轨电车的微观模拟。为了进行准确的交通模拟，需要保证输入数据的质量和清理主数据。因此，对数据进行调整，以再现现实世界的场景，并将其转换为仿真模型的可读格式。车辆需求根据法兰克福交通局提供的交通统计数据进行校准。为了模拟道路交通和道路网络，分别使用重力数学模型和开放街道地图生成原点和目的地矩阵。这个过程耗时且需要努力。然而，这个过程对于获得现实的结果至关重要。下一步，使用SUMO (Simulation of Urban mobility)对道路交通进行模拟。最后，通过仿真得到电动汽车相关关键性能指标:总行程时间和总延迟时间。将真实世界的场景与五个备选场景进行比较。kpi的比较表明，与电动汽车优先级场景相比，现实场景导致的旅行时间更长。在最小的情况下，电动汽车的整体旅行时间大大减少，正如我们所知，在电动汽车的情况下，即使节省几秒钟对有需要的人来说也是至关重要的。
{"title":"Analysis and Modelling of Road Traffic Using SUMO to Optimize the Arrival Time of Emergency Vehicles","authors":"Shamli Soni, K. Weronek","doi":"10.52825/scp.v4i.225","DOIUrl":"https://doi.org/10.52825/scp.v4i.225","url":null,"abstract":"Traffic simulation tools are used by city planners and traffic professionals over the years for modelling and analysis of existing and future infrastructural or policy implementations. There are numerous studies on emergency vehicle (EV) prioritization in cities all over the world, but every area is unique and requires the data collection and simulation to be done separately. In this case, the focus area is the Mörfelder Landstraße in Frankfurt am Main, Germany, one of the busiest streets in this city. Thestudy illustrates demand modelling, simulation and evaluation of a traffic improvement strategy for EVs. Vehicular traffic such as passenger cars and trams are simulated microscopically. To perform accurate traffic simulation, input data quality assurance and cleansing of Master Data is required. Therefore, the data is adapted to reproduce the real-world scenario and transformed into the readable format for the simulation model. Vehicular demand is calibrated by traffic count data provided by the Frankfurt Traffic Department. To model road traffic and road network, origin destination matrices using the Gravity Mathematical Model and Open Street Maps are generated, respectively. This process is time-consuming and requires effort. However, this process is critical to get realistic results. In the next step, the road traffic is simulated using SUMO (Simulation of Urban mobility). Finally, EV relevant key performance indicators (KPIs): total trip time and total delay time are derived from simulations. The real-world scenario is compared with five alternative scenarios. The comparison of the KPIs revealed that the real-world scenario results in longer travel times compared to the EV-prioritization scenario. In the least case, the overall travel times for EV has decreased significantly and, as we know, in the case of EVs, even a few seconds saved could prove crucial for a person in need.","PeriodicalId":439794,"journal":{"name":"SUMO Conference Proceedings","volume":"112 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123232417","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In transportation, a vehicle's route is one of the most private information. However, to mutually learn some phenomena in a city, for example, parking lot occupancies, we might have to reveal information about it. In this paper, we focus on assessing the privacy loss in a vehicular federated machine learning system. For the analysis, we used the Monaco SUMO Traffic Scenario (MoST). We also used the simulation inputs as statistical data to calculate privacy loss metrics. Results show that a vehicular federated machine learning system may pose a smaller privacy threat than individual learning, but its performance is lower compared to a centralized learning approach. Due to the vast amount of data and processing time, we also describe a method to build a Docker image of SUMO together with a software client-server architecture for SUMO-based learning systems on multiple computers.
{"title":"SUMO Simulations for Federated Learning in Communicating Autonomous Vehicles","authors":"Levente Alekszejenkó, Tadeusz Dobrowiecki","doi":"10.52825/scp.v4i.221","DOIUrl":"https://doi.org/10.52825/scp.v4i.221","url":null,"abstract":"In transportation, a vehicle's route is one of the most private information. However, to mutually learn some phenomena in a city, for example, parking lot occupancies, we might have to reveal information about it. In this paper, we focus on assessing the privacy loss in a vehicular federated machine learning system. For the analysis, we used the Monaco SUMO Traffic Scenario (MoST). We also used the simulation inputs as statistical data to calculate privacy loss metrics. Results show that a vehicular federated machine learning system may pose a smaller privacy threat than individual learning, but its performance is lower compared to a centralized learning approach. \u0000Due to the vast amount of data and processing time, we also describe a method to build a Docker image of SUMO together with a software client-server architecture for SUMO-based learning systems on multiple computers.","PeriodicalId":439794,"journal":{"name":"SUMO Conference Proceedings","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134242896","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Travel demand is an essential input for the creation of traffic models. However, estimating travel demand to accurately represent traffic behaviour usually requires the collection of extensive sets of data on traffic behaviour. Traffic counts are a comparably cost effective and reproducible source of information on travel demand. The utilisation of traffic counts to estimate demand is commonly found in the literature as the static and dynamic O-D estimation problem. A variety of approaches have been developed over recent decades to tackle this problem. Usually initial estimates of the O-D matrix are calibrated by utilising traffic counts and considering different assignment models. Other approaches for the estimation of travel demand solely based on traffic measurements can be found in the simulation software SUMO. The present work demonstrates the systematic development of a network model in SUMO in the inner city of Munich. In a sample network the estimation of travel demand through the tools flowrouter and routeSampler is tested by utilising flow measurements from induction loop detectors. The tests delivered unsatisfactory results, which is proven through observations of traffic flows in the resulting simulations as well as comparisons to historic traffic counts. The lack of sufficient detector data and the complexity of the sample network are discussed as the main reasons for the results. It is concluded that the applied tools should be tested in future studies with a more extensive dataset to perform a more comprehensive review of both tools. Therefore, we deliver specific requirements based on the network example of Munich.
{"title":"Calibration of a Microscopic Traffic Simulation in an Urban Scenario Using Loop Detector Data","authors":"A. Keler, A. Kunz, S. Amini, K. Bogenberger","doi":"10.52825/scp.v4i.223","DOIUrl":"https://doi.org/10.52825/scp.v4i.223","url":null,"abstract":"Travel demand is an essential input for the creation of traffic models. However, estimating travel demand to accurately represent traffic behaviour usually requires the collection of extensive sets of data on traffic behaviour. Traffic counts are a comparably cost effective and reproducible source of information on travel demand. The utilisation of traffic counts to estimate demand is commonly found in the literature as the static and dynamic O-D estimation problem. A variety of approaches have been developed over recent decades to tackle this problem. Usually initial estimates of the O-D matrix are calibrated by utilising traffic counts and considering different assignment models. Other approaches for the estimation of travel demand solely based on traffic measurements can be found in the simulation software SUMO. The present work demonstrates the systematic development of a network model in SUMO in the inner city of Munich. In a sample network the estimation of travel demand through the tools flowrouter and routeSampler is tested by utilising flow measurements from induction loop detectors. The tests delivered unsatisfactory results, which is proven through observations of traffic flows in the resulting simulations as well as comparisons to historic traffic counts. The lack of sufficient detector data and the complexity of the sample network are discussed as the main reasons for the results. It is concluded that the applied tools should be tested in future studies with a more extensive dataset to perform a more comprehensive review of both tools. Therefore, we deliver specific requirements based on the network example of Munich.","PeriodicalId":439794,"journal":{"name":"SUMO Conference Proceedings","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128218447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Deep Reinforcement Learning (DRL) is a promising data-driven approach for traffic signal control, especially because DRL can learn to adapt to varying traffic demands. For that, DRL agents maximize a scalar reward by interacting with an environment. However, one needs to formulate a suitable reward, aligning agent behavior and user objectives, which is an open research problem. We investigate this problem in the context of traffic signal control with the objective of minimizing CO2 emissions at intersections. Because CO2 emissions can be affected by multiple factors outside the agent’s control, it is unclear if an emission-based metric works well as a reward, or if a proxy reward is needed. To obtain a suitable reward, we evaluate various rewards and combinations of rewards. For each reward, we train a Deep Q-Network (DQN) on homogeneous and heterogeneous traffic scenarios. We use the SUMO (Simulation of Urban MObility) simulator and its default emission model to monitor the agent’s performance on the specified rewards and CO2 emission. Our experiments show that a CO2 emission-based reward is inefficient for training a DQN, the agent’s performance is sensitive to variations in the parameters of combined rewards, and some reward formulations do not work equally well in different scenarios. Based on these results, we identify desirable reward properties that have implications to reward design for reinforcement learning-based traffic signal control.
深度强化学习(DRL)是一种很有前途的数据驱动交通信号控制方法，特别是因为DRL可以学习适应不同的交通需求。为此，DRL代理通过与环境交互来最大化标量奖励。然而，人们需要制定一个合适的奖励，使代理行为和用户目标保持一致，这是一个开放的研究问题。我们在交通信号控制的背景下研究这个问题，目标是在十字路口减少二氧化碳的排放。由于二氧化碳排放会受到多个因素的影响，而这些因素是不受行为主体控制的，因此尚不清楚基于排放的指标是否能很好地作为一种奖励，或者是否需要一种代理奖励。为了获得合适的奖励，我们评估各种奖励和奖励组合。对于每个奖励，我们在同质和异构流量场景上训练深度q网络(DQN)。我们使用SUMO (Simulation of Urban MObility)模拟器及其默认排放模型来监控agent在指定奖励和CO2排放下的绩效。我们的实验表明，基于二氧化碳排放的奖励对于训练DQN是低效的，代理的性能对组合奖励参数的变化很敏感，并且一些奖励公式在不同的场景下效果不一样。基于这些结果，我们确定了理想的奖励属性，这些属性对基于强化学习的交通信号控制的奖励设计有影响。
{"title":"Challenges in Reward Design for Reinforcement Learning-based Traffic Signal Control: An Investigation using a CO2 Emission Objective","authors":"Max Schumacher, C. Adriano, H. Giese","doi":"10.52825/scp.v4i.222","DOIUrl":"https://doi.org/10.52825/scp.v4i.222","url":null,"abstract":"\u0000 \u0000 \u0000Deep Reinforcement Learning (DRL) is a promising data-driven approach for traffic signal control, especially because DRL can learn to adapt to varying traffic demands. For that, DRL agents maximize a scalar reward by interacting with an environment. However, one needs to formulate a suitable reward, aligning agent behavior and user objectives, which is an open research problem. We investigate this problem in the context of traffic signal control with the objective of minimizing CO2 emissions at intersections. Because CO2 emissions can be affected by multiple factors outside the agent’s control, it is unclear if an emission-based metric works well as a reward, or if a proxy reward is needed. To obtain a suitable reward, we evaluate various rewards and combinations of rewards. For each reward, we train a Deep Q-Network (DQN) on homogeneous and heterogeneous traffic scenarios. We use the SUMO (Simulation of Urban MObility) simulator and its default emission model to monitor the agent’s performance on the specified rewards and CO2 emission. Our experiments show that a CO2 emission-based reward is inefficient for training a DQN, the agent’s performance is sensitive to variations in the parameters of combined rewards, and some reward formulations do not work equally well in different scenarios. Based on these results, we identify desirable reward properties that have implications to reward design for reinforcement learning-based traffic signal control. \u0000 \u0000 \u0000","PeriodicalId":439794,"journal":{"name":"SUMO Conference Proceedings","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125785389","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：