Um melhor uso da infraestrutura da rede de transporte é um ponto fundamental para atenuar os efeitos dos congestionamentos no trânsito. Este trabalho utiliza aprendizado por reforço multiagente (MARL) para melhorar o uso da infraestrutura e, consequentemente, mitigar tais congestionamentos. A partir disso, diversos desafios surgem. Primeiro, a maioria da literatura assume que os motoristas aprendem (semáforos não possuem nenhum tipo de aprendizado) ou os semáforos aprendem (motoristas não alteram seus comportamentos). Em segundo lugar, independentemente do tipo de classe de agentes e do tipo de aprendizado, as ações são altamente acopladas, tornando a tarefa de aprendizado mais difícil. Terceiro, quando duas classes de agentes co-aprendem, as tarefas de aprendizado de cada agente são de natureza diferente (do ponto de vista do aprendizado por reforço multiagente). Finalmente, é utilizada uma modelagem microscópica, que modela os agentes com um alto nível de detalhes, o que não é trivial, pois cada agente tem seu próprio ritmo de aprendizado. Portanto, este trabalho não propõe somente a abordagem de co-aprendizado em agentes que atuam em ambiente compartilhado, mas também argumenta que essa tarefa precisa ser formulada de forma assíncrona. Além disso, os agentes motoristas podem atualizar os valores das ações disponíveis ao receber informações de outros motoristas. Os resultados mostram que a abordagem proposta, baseada no coaprendizado, supera outras políticas em termos de tempo médio de viagem. Além disso, quando o co-aprendizado é utilizado, as filas de veículos parados nos semáforos são menores. / A better use of transport network infrastructure is a key point in mitigating the effects of traffic congestion. This work uses multiagent reinforcement learning (MARL) to improve the use of infrastructure and, consequently, to reduce such congestion. From this, several challenges arise. First, most literature assumes that drivers learn (traffic lights do not have any type of learning) or the traffic lights learn (drivers do not change their behaviors). Second, regardless of the type of agent class and the type of learning, the actions are highly coupled, making the learning task more difficult. Third, when two classes of agents co-learn, the learning tasks of each agent are of a different nature (from the point of view of multiagent reinforcement learning). Finally, a microscopic modeling is used, which models the agents with a high level of detail, which is not trivial, since each agent has its own learning pace. Therefore, this work does not only propose the co-learnig approach in agents that act in a shared environment, but also argues that this taks needs to be formulated asynchronously. In addtion, driver agents can update the value of the available actions by receiving information from other drivers. The results show that the proposed approach, based on co-learning, outperforms other policies regarding average travel time. Also, when co-learning is use, queues of stopped vehicles at traffic lights are lower.

Os planos de coordenação semafórica são reconhecidamente eficientes para promover a fluidez das correntes de tráfego e melhorar a qualidade operacional do sistema viário. Portanto, os técnicos responsáveis pela gestão do tráfego deveriam, sempre que possível, adotar métodos para a definição adequada das defasagens, utilizando ferramentas que auxiliem na decisão das operações. Porém, uma pesquisa realizada em cidades das regiões Sul e Sudeste sobre as estratégias de coordenação empregadas, demonstrou especialmente para municípios de porte médio que somente 13% utilizam ferramentas computacionais para prover planos de coordenação nos semáforos monitorados por centrais. Para os demais corredores semaforizados, cerca de 27% utilizam o diagrama espaço-tempo, 36% realizam ajustes locais através da observação do tráfego, 14% utilizam veículo-teste e 23% não adotam esquemas de coordenação. O objetivo desta dissertação é avaliar a eficiência de dois programas de coordenação semafórica que poderiam ser utilizados em cidades que não dispõem de técnicas mais eficientes para definir os planos semafóricos. Uma revisão bibliográfica sobre os métodos mais conhecidos foi realizada, identificando que os programas disponíveis seguem três critérios distintos: a maximização da largura da banda verde; a minimização dos atrasos e paradas; e a combinação das vantagens de ambos os critérios. O primeiro programa, cujo nome é SBAND, consiste em uma implementação computacional do critério de maximização de banda baseada no método half-integer synchronization, proposto por Morgan e Little. O segundo programa é o simulador INTEGRATION, que coordena semáforos baseado no critério de minimização de atrasos e paradas. Com esses programas, foram gerados planos semafóricos para diversas condições de tráfego em vias arteriais de Londrina e São Carlos, simuladas com o INTEGRATION. A análise das medidas de desempenho - tempo de percurso, atraso médio e número médio de paradas - indica que é possível planos semafóricos melhores do que aqueles em vigência nos corredores analisados, com redução dos valores médios destas medidas. A análise sugere que o programa INTEGRATION é mais recomendado para os casos em que o fluxo de veículos é elevado e semelhante na via principal e nas transversais, enquanto o programa SBAND é recomendado quando a via principal apresenta volume de tráfego leve ou moderado, maior do que nas vias secundárias. / Traffic signal synchronization strategies are useful to improve traffic flow mobility and level of service of arterial streets. Therefore, technicians in charge of management of the traffic system operation should take advantage of the benefits of these tools, as always as possible, in order to define the most adequate traffic signal offsets. However, interviews carried on mid-size cities of southern and southwestern regions of Brazil indicated that only 13% of these cities define plans for traffic signal networks, controlled by a central control room, based on any type of traffic signal coordination software. For non-controlled streets, 27% of the cities define offsets using the traditional space-time diagram, 36% make local adjustments in a trial and error basis, 14% use a test vehicle and the remaining 23% do not use any strategy for synchronization. Based on these findings, the objective of this master thesis is to assess the performance of two traffic signal synchronization tools that might be used in cities which do not make use of more efficient techniques to define signal plans. Literature review indicates that the available softwares are developed according to three distinct approaches: bandwidth maximization, minimization of delays and stops and combination of both previous methods. The first software tested, named SBAND, is based on the Morgan and Little half-integer synchronization algorithm, which maximizes bandwidth of signalized arterial streets. The second software is the INTEGRATION traffic simulation tool, which synchronizes traffic signals minimizing delays and stops. Different signal plans were defined by means of these two softwares, for several traffic conditions in arterial streets of Londrina and São Carlos, and simulated with INTEGRATION. The analysis of measures of effectiveness generated on the simulations – travel time, average delay and average number of stops – indicate that is possible to obtain better traffic signal plans, with reduction on the measures of effectiveness for the plans currently adopted for these arterials. The results also suggest that INTEGRATION is more recommended for scenarios when traffic flow on principal arterials is heavy and similar to the traffic on secondary streets, while SBAND is recommended for light to moderate traffic flow on main street, with through trips predominating and lighter traffic on secondary streets.

Traffic simulation and signal timing optimization are classified in structure into two main categories: (i) Macroscopic or Microscopic; (ii) Deterministic or Stochastic. Performance of the optimized signal timing derived by any tool is influenced by the methodology used in how calculations are executed in a particular tool. In this study, the performance of the optimal signal timing plans developed by two of the most popular traffic analysis tools, HCS and Tru-Traffic, each of them has its inbuilt objective function(s) to optimize signal timing for intersection, is compared with an ideal and an existing timing plans (base case) for the area of study using the microsimulation software VISSIM. An urban arterial with 29 intersections and high traffic in Fort Lauderdale, Florida serves as the test bed. To eliminate unfair superiority in the results, all experiments were performed under identical geometry and traffic conditions in each tool. Comparison of the optimized plans is conducted on the basis of average delay, average stopped delay, average number of stops, number of vehicles completed trips, latent delay, and latent demand from the simulated vehicle network performance evaluation results in VISSIM. The results indicate that, overall, HCS with its overall delay objective and the Tru-Traffic programs produce signal timing with comparable quality that performed similar to the un-optimized base case for most of the performance measures. / Includes bibliography. / Thesis (M.S.)--Florida Atlantic University, 2019. / FAU Electronic Theses and Dissertations Collection

Signal retiming, or signal optimization process, has not changed much over the last few decades. Traditional procedures rely on low-resolution data and a low-fidelity modeling approach. Such developed signal timing plans always require a fine-tuning process for deployed signal plans in field, thus questioning the very benefits of signal optimization. New trends suggest the use of high-resolution data, which are not easily available. At the same time, many improvements could be made if the traditional signal retiming process was modified to include the use of medium-resolution data and high-fidelity modeling. This study covers such an approach, where a traditional retiming procedure is modified to utilize large medium-resolution data sets, high-fidelity simulation models, and powerful stochastic optimization to develop robust signal timing plans. The study covers a 28-intersection urban corridor in Southeastern Florida. Medium-resolution data are used to identify peak-hour, Day-Of-Year (DOY) representative volumes for major seasons. Both low-fidelity and high-fidelity models are developed and calibrated with high precision to match the field signal operations. Then, by using traditional and stochastic optimization tools, signal timing plans are developed and tested in microsimulation. The findings reveal shortcomings of the traditional approach. Signal timing plans developed from medium-resolution data and high-fidelity modeling approach reduce average delay by 5%-26%. Travel times on the corridor are usually reduced by up to 10.5%, and the final solution does not transfer delay on the other neighboring streets (illustrated through latent delay), which is also decreased by 10%-49% when compared with the traditional results. In general, the novel approach has shown a great potential. The next step should be field testing and validation. / Includes bibliography. / Thesis (M.S.)--Florida Atlantic University, 2019. / FAU Electronic Theses and Dissertations Collection

The purpose of this thesis was to perform microsimulation analyses on intersections near Highway-Railroad Grade Crossings (HRGCs) to determine if controlling mean train speed and train speed variability would improve safety and reduce delays. This research focused on three specific areas. First, average vehicle delay was examined, and this delay was compared for seven specific train speed distributions, including existing conditions. Furthermore, each distribution was associated with train detectors that were placed at the distance the fastest train could travel during the given warning time. Second, pedestrian cutoffs were investigated. These cutoffs represented an occasion when the pedestrian phases were truncated or shortened due to railroad signal preemption. Finally, vehicle emissions were analyzed using a modal emissions model. A microscopic simulation model of the Wellborn Corridor in College Station, Texas was created using VISSIM. The model was run twenty times in each train speed distribution for each of three train lengths. Average vehicle delay was collected for three intersections, and delays were compared using the Pooled t-test with a 95% confidence interval. Comparisons were made between the distributions, and generally, distributions with higher mean train speeds were associated with lower average delay, and train length was not a significant factor. Unfortunately, pedestrian cutoffs were not specifically controlled in this project; therefore, no statistical conclusions can be made with respect to the pedestrian cutoff problem. However, example cases were devised to demonstrate how these cutoffs could be avoided. In addition, vehicle emissions were examined using the vehicle data from VISSIM as inputs for CMEM (Comprehensive Modal Emissions Model). For individual vehicles, as power (defined as the product of velocity and acceleration) increased, emissions increased. When comparing emissions from different train speed distributions, few significant differences were found. However, a scenario with no train was tested, and it was shown to have significantly higher emissions than three of the distributions with trains. Ultimately, this thesis shows that average vehicle delay and vehicle emissions could be lowered by specific train speed distributions. Also, work could be done to investigate the pedestrian cutoff problem.

Neue Methoden zur Verkehrsdatenerfassung wie die Fahrzeug-Infrastruktur-Kommunikation, der Floating Car-Ansatz und die Videodetektion eröffnen die Möglichkeit, neue Verfahren zur verkehrsabhängigen Lichtsignalanlagensteuerung zu realisieren. In dem Beitrag wird ein Verfahren beschrieben, das aus diesen Quellen Daten in Form von Fahrzeugverlustzeiten direkt zur Steuerung eines Einzelknotens verwendet. Die robuste Ausgestaltung des Verfahrens sorgt dabei dafür, dass auch mit einer lückenhaften Datenlage, wie z. B. aufgrund geringer Ausstattungsraten kommunikationsfähiger Fahrzeuge, angemessen umgegangen werden kann. Mit Hilfe einer mikroskopischen Simulationsstudie wird nachgewiesen, dass das neue Verfahren bei der Qualität des Verkehrsablaufs das gleiche Niveau wie eine traditionelle Zeitlückensteuerung erreicht oder dieses unter bestimmten Bedingungen sogar übersteigt. Mit abnehmender Ausstattungsrate ergibt sich dabei allerdings ein Qualitätsverlust, der ebenfalls mit Hilfe der mikroskopischen Simulation quantifiziert wird und wichtige Erkenntnisse für einen möglichen Praxistest liefert. / State-of-the-art traffic data sources like Car-to-Infrastructure communication, Floating Car Data and video detection offer great new prospects for vehicle-actuated traffic signal control. Due to this, the article deals with a recent approach which uses vehicles’ delay times for real-time control of traffic signals at an isolated intersection. One of the strengths of the new approach is that it can handle also incomplete data sets, e.g. caused by low penetration rates of vehicles equipped with Car-to-Infrastructure communication technology, in an appropriate manner. Based on a microscopic simulation study the high quality of this innovative approach is demonstrated, which is equal or even outperforms the well-known headway-based control. However, a decreasing penetration rate of equipped vehicles means a reduced quality of signals’ control, which is quantified in the microscopic simulation study, too, and provides useful information for tests in the field.

Analytical Model of Traffic Signal Structures (TSS) is developed based on a continuous system method to observe dynamic characteristics of the structures. Conventional and basic continuous system method can show the approximate dynamic characteristics of the TSS, but the discretized continuous analytical model is proposed to get more accurate and realistic results of the TSS. In addition, the discretized continuous model can alternatively analyze the effect of the tapered cross-sectional members which are real model of TSS. For the verification of the analytical model, the dynamic characteristics of the numerical solutions by modal analysis in ABAQUS and the results of experimental measurements are provided. Compared with the numerical solutions and the experimental results, the analytical solution for each member shows its considerable accuracy. In addition, it will be also able to accurately express the effects of the linearly tapered cross-sectional member with more discretized continuous structural system. Moreover, the discretized analytical model of the TSS has the usability to observe the effects of boundary flexibility.

The purpose of this thesis was to perform microsimulation analyses on intersections near Highway-Railroad Grade Crossings (HRGCs) to determine if controlling mean train speed and train speed variability would improve safety and reduce delays. This research focused on three specific areas. First, average vehicle delay was examined, and this delay was compared for seven specific train speed distributions, including existing conditions. Furthermore, each distribution was associated with train detectors that were placed at the distance the fastest train could travel during the given warning time. Second, pedestrian cutoffs were investigated. These cutoffs represented an occasion when the pedestrian phases were truncated or shortened due to railroad signal preemption. Finally, vehicle emissions were analyzed using a modal emissions model. A microscopic simulation model of the Wellborn Corridor in College Station, Texas was created using VISSIM. The model was run twenty times in each train speed distribution for each of three train lengths. Average vehicle delay was collected for three intersections, and delays were compared using the Pooled t-test with a 95% confidence interval. Comparisons were made between the distributions, and generally, distributions with higher mean train speeds were associated with lower average delay, and train length was not a significant factor. Unfortunately, pedestrian cutoffs were not specifically controlled in this project; therefore, no statistical conclusions can be made with respect to the pedestrian cutoff problem. However, example cases were devised to demonstrate how these cutoffs could be avoided. In addition, vehicle emissions were examined using the vehicle data from VISSIM as inputs for CMEM (Comprehensive Modal Emissions Model). For individual vehicles, as power (defined as the product of velocity and acceleration) increased, emissions increased. When comparing emissions from different train speed distributions, few significant differences were found. However, a scenario with no train was tested, and it was shown to have significantly higher emissions than three of the distributions with trains. Ultimately, this thesis shows that average vehicle delay and vehicle emissions could be lowered by specific train speed distributions. Also, work could be done to investigate the pedestrian cutoff problem.

Signal control systems have experienced tremendous development both in hardware and in control strategies in the past 50 years since the advent of the first electronic traffic signal control device. The state-of-art real-time signal control strategies rely heavily on infrastructure-based sensors, including in-pavement or video based loop detectors for data collection. With the emergence of connected vehicle technology, mobility applications utilizing vehicle to infrastructure (V2I) communications enable the intersection to acquire a much more complete picture of the nearby vehicle states. Based on this new source of data, traffic controllers should be able to make "smarter" decisions. This dissertation investigates the traffic signal control strategies in a connected vehicle environment considering mobility as well as safety. A system architecture for connected vehicle based signal control applications under both a simulation environment and in the real world has been developed. The proposed architecture can be applied to applications such as adaptive signal control, signal priority including transit signal priority (TSP), freight signal priority (FSP), emergency vehicle preemption, and integration of adaptive signal control and signal priority. Within the framework, the trajectory awareness of connected vehicles component processes and stores the connected vehicle data from Basic Safety Message (BSM). A lane level intersection map that represents the geometric structure was developed. Combined with the map and vehicle information from BSMs, the connected vehicles can be located on the map. Some important questions specific to connected vehicle are addressed in this component. A geo-fencing area makes sure the roadside equipment (RSE) receives the BSM from only vehicles on the roadway and within the Dedicated Short-range Communications (DSRC) range. A mechanism to maintain anonymity of vehicle trajectories to ensure privacy is also developed. Vehicle data from the trajectory awareness of connected vehicles component can be used as the input to a real-time phase allocation algorithm that considers the mobility aspect of the intersection operations. The phase allocation algorithm applies a two-level optimization scheme based on the dual ring controller in which phase sequence and duration are optimized simultaneously. Two objective functions are considered: minimization of total vehicle delay and minimization of queue length. Due to the low penetration rate of the connected vehicles, an algorithm that estimates the states of unequipped vehicles based on connected vehicle data is developed to construct a complete arrival table for the phase allocation algorithm. A real-world intersection is modeled in VISSIM to validate the algorithms. Dangerous driving behaviors may occur if a vehicle is trapped in the dilemma zone which represents one safety aspect of signalized intersection operation. An analytical model for estimating the number of vehicles in dilemma zone (NVDZ) is developed on the basis of signal timing, arterial geometry, traffic demand, and driving characteristics. The analytical model of NVDZ calculation is integrated into the signal optimization to perform dilemma zone protection. Delay and NVDZ are formulated as a multi-objective optimization problem addressing efficiency and safety together. Examples show that delay and NVDZ are competing objectives and cannot be optimized simultaneously. An economic model is applied to find the minimum combined cost of the two objectives using a monetized objective function. In the connected vehicle environment, the NVDZ can be calculated from connected vehicle data and dilemma zone protection is integrated into the phase allocation algorithm. Due to the complex nature of traffic control systems, it is desirable to utilize traffic simulation in order to test and evaluate the effectiveness and safety of new models before implementing them in the field. Therefore, developing such a simulation platform is very important. This dissertation proposes a simulation environment that can be applied to different connected vehicle related signal control applications in VISSIM. Both hardware-in-the-loop (HIL) and software-in-the-loop (SIL) simulation are used. The simulation environment tries to mimic the real world complexity and follows the Society of Automotive Engineers (SAE) J2735 standard DSRC messaging so that models and algorithms tested in the simulation can be directly applied in the field with minimal modification. Comprehensive testing and evaluation of the proposed models are conducted in two simulation networks and one field intersection. Traffic signal priority is an operational strategy to apply special signal timings to reduce the delay of certain types of vehicles. The common way of serving signal priority is based on the "first come first serve" rule which may not be optimal in terms of total priority delay. A priority system that can serve multiple requests with different priority levels should perform better than the current method. Traditionally, coordination is treated in a different framework from signal priority. However, the objectives of coordination and signal priority are similar. In this dissertation, adaptive signal control, signal priority and coordination are integrated into a unified framework. The signal priority algorithm generates a feasible set of optimal signal schedules that minimize the priority delay. The phase allocation algorithm considers the set as additional constraints and tries to minimize the total regular vehicle delay within the set. Different test scenarios including coordination request, priority vehicle request and combination of coordination and priority requests are developed and tested.

Thesis (Ph. D.)--Michigan State University. Dept. of Civil Engineering, 2008. / Title from PDF t.p. (viewed on July 7, 2009) Includes bibliographical references (p. 196-201). Also issued in print.