Assunto bastante abordado quando se trata de Sistemas Inteligentes de Transportes (ITS), a identificação veicular - utilizada em grande parte das aplicações de ITS deve ser entendida como um conjunto de recursos de hardware, software e telecomunicações, que interagem para atingir, do ponto de vista funcional, o objetivo de, conseguir extrair e transmitir, digitalmente, a identidade de um veículo. É feita tanto por sistemas que transmitem e recebem uma identidade digital quanto por sistemas que, instalados na infraestrutura da via, são capazes de reconhecer a placa dos veículos circulantes. Quando se trata da identificação automática por meio do reconhecimento da placa veicular, os estudos têm se concentrado sobremaneira nas tecnologias de processamento de imagens, não abordando - em sua maioria - uma visão sistêmica, necessária para compreender de maneira mais abrangente todas as variáveis que podem interferir na eficácia da identificação. Com o objetivo de contribuir para melhor entender e utilizar os sistemas de reconhecimento automático de placas veiculares, este trabalho propõe um modelo sistêmico, em camadas, para representar seus componentes. Associada a esse modelo, propõe uma classificação para os diversos tipos de falhas que podem prejudicar seu desempenho. Uma análise desenvolvida com resultados obtidos em testes realizados em campo com sistemas de identificação de placas voltados à fiscalização de veículos aponta resultados relevantes e limitações para obter correlações entre variáveis, em função dos diversos fatores que podem influenciar os resultados. Algumas entrevistas realizadas apontam os tipos de falhas que ocorrem com mais frequência durante a operação desses sistemas. Finalmente, este trabalho propõe futuros estudos e apresenta um glossário de termos, que poderá ser útil a novos pesquisadores. / The automatic vehicle identification is an important feature of Intelligent Transportation Systems (ITS) and is used in most ITS applications. The identification process is comprised of a group of interacting resources that involves hardware, software and telecommunication to, digitally, extract and transmit the identity of vehicles. At least two technologies may be used in the vehicle identification process: on-board devices transmitting a digital identity or systems installed on the road infrastructure, which identify and read the vehicle license plate. As far as vehicle license plate recognition is concerned, studies have been greatly focused on image processing technologies and have not addressed the problem in a systemic approach, which is very important for understanding all variables that can interfere with the effectiveness of identification. Having this approach in mind and intending to contribute for a better performance, this paper proposes a layer model representation of those systems as well as a failure type classification associated with it. An analysis, based on a significant set of results obtained from field tests of systems with plate recognition capabilities for law enforcement, shows important results as well as limitations to obtain mathematical correlation of variables. Interviews conducted with supply actors of such systems in Brazil point out the most significant sources of failures that occur during operation. Finally, the text presents potential topics for research and organizes a glossary of terms that may be useful to future researchers.

Dans un proche futur, les véhicules connectés et autonomes remplaceront nos véhicules actuels, et il sera nécessaire de repenser intégralement la mobilité. Le conducteur, avec ses lacunes, sera de plus en plus assisté, et un jour détrôné par un système embarqué, capable d'agir plus rapidement, tout en ayant une représentation plus précise et fiable de son environnement de navigation. Pourtant, il reste encore du chemin à parcourir avant d'arriver à un tel stade de maturité : l'environnement du véhicule est complexe, imprévisible et conflictuel, car partagé avec d'autres acteurs de la mobilité. Ce travail de thèse vise à anticiper l'arrivée de ces nouveaux véhicules afin de proposer des comportements coopératifs au niveau des véhicules et de l'infrastructure tout en permettant un contrôle décentralisé de ce type de système complexe. Dans le cadre d'une approche multi-agents et d'une architecture distribuée, nous proposons d'abord une modélisation par couplage des dynamiques physique et communicationnelle, auxquelles s'ajoute une intégration de la fiabilité de l'information (confiance). L'étape suivante a été de développer un cadre de simulation propice à l'implémentation de nos modèles dans le cas des véhicules connectés. Nous introduisons un nouveau simulateur de trafic, construit comme une extension d'une plateforme existante, au sein duquel les flux d'informations entre les véhicules et avec l'infrastructure prennent la forme d'échanges de messages. Toutes les informations du système proviennent de capteurs, et toutes les entités, modélisées comme des agents, sont autonomes dans leur prise de décision. De nouvelles formes de contrôle sont désormais envisageables en utilisant des consignes transmises par l'infrastructure communicante. Le couplage des dynamiques assure la cohérence et l'inter-dépendance des différents modèles dans le simulateur. Nous montrons en simulation que, grâce à l'intégration d'informations supplémentaires via la communication, les véhicules, modélisés par un modèle microscopique multi-anticipatif bilatéral, sont capables de réduire l'effet de perturbations propagées au sein d'un flux. En termes de stratégies de contrôle, une des problématiques principales est de garantir une forme de contrôle qui s'adaptera aux différentes phases de déploiement des systèmes coopératifs. L'analogie avec des problèmes de l'IA (problème de cognition) nous a mené à traiter le problème de manière plus abstraite : comment permettre à un système autonome de contrôler son environnement. Les approches constructivistes, que nous avons retenues, modélisent le processus de cognition comme un phénomène de construction itératif. Pour le trafic coopératif, l'avantage est de disposer d'un système capable de générer ses propres stratégies, en utilisant ou non des connaissances expertes, et de les faire évoluer au cours du temps pour s'adapter aux véhicules composant le flux. Les résultats de notre approche sont présentés dans deux cadres de simulation. Le premier est un prototype visant à illustrer les comportements de bas niveau dans un environnement simplifié. Nous montrons que le modèle est capable dans ce cadre de combiner différentes représentations individuelles pour construire une représentation et de s'adapter à différents contextes en les recombinant dynamiquement. Puis, dans le cadre de simulation du trafic coopératif, les résultats laissent entrevoir le potentiel de notre approche dans des applications réelles / In a near future, connected and automated vehicles will progressively replace current vehicles, leading to deep changes in transportation. The driver will be soon assisted and then replaced by an embedded system, able to act quicker, relying on a more robust and precise representation of its surrounding environment. However, some steps are still needed before coming up with such a level of automation since the vehicle environment is complex and unpredictable. This work intends to anticipate the introduction of these new kinds of vehicles by providing cooperative behaviors at both infrastructure and vehicle levels, at the same time allowing a decentralized control of these systems. We propose a distributed modeling framework, using multi-agent systems, relying on the coupling of the system dynamics: information, communication and reliability (modeled through the concept of trust). The next step was to develop a simulation framework enabling the implementation of our models for connected vehicles applications. We present a new microscopic traffic simulator, built as an extension of an existing platform, and able to model information exchanges using messages between vehicles and with the infrastructure. All data are provided by sensors and all entities, modeled as agents, are autonomous regarding their decision process. Thanks to the simulator, it is possible to imagine new control strategies relying on recommendations disseminated by the connected infrastructure. Consistency and interdependence of the simulator components are ensured by the dynamic coupling. As for the vehicles’ dynamics, we propose a bilateral multi-anticipative model that integrates additional information from communications in the vehicle decision process. Results in simulation confirm that the model is able to reduce the propagation of perturbation through the flow, leading to a more homogeneous and stable traffic. One of the major issues regarding traffic control strategies will be to dynamically adapt the action policy to the several deployment stages of cooperative transportation systems. The similarities with Artificial Intelligence problems like cognition motivate a more abstract study: how to model an autonomous system able to control its environment. We choose the constructivist approaches, that propose to model the cognition process as an iterative building process. For cooperative traffic, the benefits lie in the ability of the system to generate its own strategies, relying or not on domain specific knowledge, and then make them evolve to be adapted to vehicles in the flow. The results from our approach are presented in two distinct simulation frameworks. The first one is an experimentation prototype aiming at highlighting the low-level behaviors in a simplified environment. In this context, we show that the model is able to combine efficiently several individual concurrent representations in order to build a high-level representation that can be adapted to several contexts. The second framework is the traffic simulator where the results lead to some insights about the potential of our approach for such realistic applications

Submitted by Daniella Santos (daniella.santos@ufma.br) on 2017-11-22T16:02:27Z No. of bitstreams: 1 HeygonAraújo.pdf: 4678551 bytes, checksum: 923410aa48c69f8cf3b40b930eaa71fb (MD5) / Made available in DSpace on 2017-11-22T16:02:27Z (GMT). No. of bitstreams: 1 HeygonAraújo.pdf: 4678551 bytes, checksum: 923410aa48c69f8cf3b40b930eaa71fb (MD5) Previous issue date: 2017-07-13 / The definition of the best sequence on route duplication of freight systems consists of a complex NP-hard problem. There exists a huge variety of meta-heuristics (MH) capable of generating satisfactory solutions. However, it is fastidious to know which MH will produce the best solution for a Duplication Sequence Problem (DSP). This paper proposes a process development methodology which guides to evaluate the best duplication sequence comparing the MH’s performance with existing approaches such as linear analytical method (LAM), and thus to ensure that the system has its capacity maximized in the possible shortest time interval. If this sequence is prioritized incorrectly, it tends to generate wastes such as time and currency in new routes which will not add system capacity in short term. The potential of this methodology is demonstrated by a case study in railways. / A definição da sequência ótima de duplicação em vias de sistemas de transportes de cargas consiste de um problema de complexidade intratável. Existem uma grande variedade de Meta-heurísticas (MHs) capazes de gerar soluções satisfatórias. Entretanto, é fastidioso conhecer a MH que produzirá a melhor solução para um dado Problema de Sequenciamento de Duplicação (PSD) . Não há na literatura uma metodologia para estruturar, planejar e controlar algoritmos e processos na modelagem das variedade de MHs aplicadas para encontrar a solução ótima neste tipo de problema. Este trabalho apresenta uma metodologia de desenvolvimento de processo o qual busca pela sequência ótima de duplicação em um dado sistema de transporte comparando a performance de MH com abordagens existentes, tais como método analítico lineares (MALs), e desta forma garantir que o sistema tenha sua capacidade maximizada no menor intervalo de tempo possível. Caso esta sequência seja priorizada de forma incorreta, a tendência será o desperdício de tempo e dinheiro em novas vias as quais não agrerarão capacidade ao sistema no curto prazo. O potencial desta metodologia é demonstrado através de um estudo de caso em ferrovias.

Ramp metering is an effective management strategy, which helps to keep traffic density below the critical value, preventing breakdowns and thus maintaining the full capacity of the freeway. Warrants for ramp metering installation have been developed by a number of states around the nation. These warrants are generally simple and are based on the traffic, geometry, and safety conditions in the immediate vicinity of each ramp (local conditions). However, advanced applications of ramp metering utilize system-based metering algorithms that involve metering a number of on-ramps to address system bottleneck locations. These algorithms have been proven to perform better compared to local ramp metering algorithms. This has created a disconnection between existing agency metering warrants to install the meters and the subsequent management and operations of the ramp metering. Moreover, the existing local warrants only consider recurrent conditions to justify ramp metering installation with no consideration of the benefits of metering during non-recurrent events such as incidents and adverse weather. This dissertation proposed a methodology to identify the ramps to meter based on system-wide recurrent and non-recurrent traffic conditions. The methodology incorporates the stochastic nature of the demand and capacity and the impacts of incidents and weather using Monte Carlo simulation and a ramp selection procedure based on a linear programming formulation. The results of the Monte Carlo simulation are demand and capacity values that are used as inputs to the linear programming formulation to identify the ramps to be metered for each of the Monte Carlo experiments. This method allows the identification of the minimum number of ramps that need to be metered to keep the flows below capacities on the freeway mainline segment, while keeping the on-ramp queues from spilling back to the upstream arterial street segments. The methodology can be used in conjunction with the existing local warrants to identify the ramps that need to be metered. In addition, it can be used in benefit-cost analyses of ramp metering deployments and associated decisions, such as which ramps to meter and when to activate in real-time. The methodology is extended to address incidents and rainfall events, which result in non-recurrent congestion. For this purpose, the impacts of non-recurrent events on capacity and demand distributions are incorporated in the methodology.

This thesis focuses on two important problems related to DDDAS: interest management (data distribution) and prediction models. In order to reduce communication overhead, we propose a new interest management mechanism for mobile peer-to-peer systems. This approach involves dividing the entire space into cells and using an efficient sorting algorithm to sort the regions in each cell. A mobile landmarking scheme is introduced to implement this sort-based scheme in mobile peer-to-peer systems. The design does not require a centralized server, but rather, every peer can become a mobile landmark node to take a server-like role to sort and match the regions. Experimental results show that the scheme has better computational efficiency for both static and dynamic matching. In order to improve communication efficiency, we present a travel time prediction model based on boosting, an important machine learning technique, and combine boosting and neural network models to increase prediction accuracy. We also explore the relationship between the accuracy of travel time prediction and the frequency of traffic data collection with the long term goal of minimizing bandwidth consumption. Several different sets of experiments are used to evaluate the effectiveness of this model. The results show that the boosting neural network model outperforms other predictors.

Changeable Message Signs (CMS) are commonly utilized by transportation agencies to inform motorists of traffic, roadway, and environmental conditions. They may be used to provide information, such as delay and alternate route guidance, in the event of an incident, construction or a roadway closure. The effectiveness of CMS in managing freeway traffic, however, is a function of many factors including the number of CMS installations, the location of CMS, the messages displayed, varied traffic network characteristics, and drivers' response to incident conditions and CMS information. The objective of this thesis is to develop a CMS location planning model that can be used by transportation agencies to develop a CMS location plan that could achieve the largest long-term benefit to the system. This research is mainly motivated by the lack of systematic, robust and practical methods for locating CMS. State-of-practice methods rely mostly on the practitioner's experience and judgement. Other methods fail to incorporate reasonable driver behaviour models, consider time-varying demand, allow for computational efficiency on large networks, or consider the spatial variation of incidents on a traffic network. A new CMS location optimization model has been developed that is unique in both model realism and computational efficiency. The model incorporates several components to estimate incident delay, predict driver response, estimate network-wide benefit, and choose those CMS locations that would provide the most benefit. Deterministic queuing methods are used in conjunction with historic incident characteristics to approximate the delay impact of an incident with and without CMS. A discrete choice model is used to predict the rate at which drivers would switch from the incident route to a less congested alternative under CMS information. A network traffic assignment model is then incorporated in an attempt to estimate the resulting traffic induced by incidents. Genetic algorithms are utilized as an optimization technique to choose a set of CMS that would provide the most benefit. An extensive computational analysis was performed on both a hypothetical network and a segment of Highway 401 through Toronto. A sensitivity analysis was performed to test the model's response to parameter and data estimation errors. The model was found to be most sensitive to the diversion model parameters. The model produced reasonable results with locations selected upstream of major freeway interchange diversion points. Considering the additional components included in the proposed model, and its ability to consider more location schemes, the proposed model may be considered superior to previous CMS location models.

Changeable Message Signs (CMS) are commonly utilized by transportation agencies to inform motorists of traffic, roadway, and environmental conditions. They may be used to provide information, such as delay and alternate route guidance, in the event of an incident, construction or a roadway closure. The effectiveness of CMS in managing freeway traffic, however, is a function of many factors including the number of CMS installations, the location of CMS, the messages displayed, varied traffic network characteristics, and drivers' response to incident conditions and CMS information. The objective of this thesis is to develop a CMS location planning model that can be used by transportation agencies to develop a CMS location plan that could achieve the largest long-term benefit to the system. This research is mainly motivated by the lack of systematic, robust and practical methods for locating CMS. State-of-practice methods rely mostly on the practitioner's experience and judgement. Other methods fail to incorporate reasonable driver behaviour models, consider time-varying demand, allow for computational efficiency on large networks, or consider the spatial variation of incidents on a traffic network. A new CMS location optimization model has been developed that is unique in both model realism and computational efficiency. The model incorporates several components to estimate incident delay, predict driver response, estimate network-wide benefit, and choose those CMS locations that would provide the most benefit. Deterministic queuing methods are used in conjunction with historic incident characteristics to approximate the delay impact of an incident with and without CMS. A discrete choice model is used to predict the rate at which drivers would switch from the incident route to a less congested alternative under CMS information. A network traffic assignment model is then incorporated in an attempt to estimate the resulting traffic induced by incidents. Genetic algorithms are utilized as an optimization technique to choose a set of CMS that would provide the most benefit. An extensive computational analysis was performed on both a hypothetical network and a segment of Highway 401 through Toronto. A sensitivity analysis was performed to test the model's response to parameter and data estimation errors. The model was found to be most sensitive to the diversion model parameters. The model produced reasonable results with locations selected upstream of major freeway interchange diversion points. Considering the additional components included in the proposed model, and its ability to consider more location schemes, the proposed model may be considered superior to previous CMS location models.

Flow of traffic on freeways and limited access highways can be represented as a series of kinemetic waves. Solutions to these systems of equations become problematic under congested traffic flow conditions, and under complicated (real-world) networks. A simplified theory of kinematics waves was previously proposed. Simplifying elements includes translation of the problem to moving coordinate system, adoption of bi-linear speed-density relationships, and adoption of restrictive constraints at the on- and off-ramps. However, these simplifying assumptions preclude application of this technique to most practical situations. This research explores the limitations of the simplified theory of kinematic waves. First this research documents a relaxation of several key constraints. In the original theory, priority was given to on-ramp merging vehicles so that they can bypass any queue at the merge. This research proposes to relax this constraint using a capacity-based weighted fair queuing (CBWFQ) merge model. In the original theory, downstream queue affects upstream traffic as a whole and exiting traffic can always be able to leave as long as it gets to the diverge. This research proposes that this diverge constraint be replaced with a contribution-based weighted splitting (CBWS) diverge model. This research proposes a revised notation system, permitting the solution techniques to be extended to freeway networks with multiple freeways and their ramps. This research proposes a generalization to permit application of the revised theory to general transportation networks. A generalized CBWFQ merge model and a generalized CBWS diverge model are formulated to deal with merging and diverging traffic. Finally, this research presents computational procedure for solving the new system of equations. Comparisons of model predictions with field observations are conducted on GA 400 in Atlanta. Investigations into the performance of the proposed CBWFQ and CBWS models are conducted. Results are quite encouraging, quantitative measures suggest satisfactory accuracy with narrow confidence interval.

This research develops and evaluates an Advanced Traveler Information System (ATIS) model using a Vehicle-to-Vehicle (V2V) communication system (referred to as the GATIS-V2V model) with the off-the-shelf microscopic simulation model, VISSIM. The GATIS-V2V model is tested on notional small traffic networks (non-signalized and signalized) and a 6X6 typical urban grid network (signalized traffic network). The GATIS-V2V model consists of three key modules: vehicle communication, on-board travel time database management, and a Dynamic Route Guidance System (DRGS). In addition, the system performance has been enhanced by applying three complementary functions: Autonomous Automatic Incident Detection (AAID), a minimum sample size algorithm, and a simple driver behavior model. To select appropriate parameter ranges for the complementary functions a sensitivity analysis has been conducted. The GATIS-V2V performance has been investigated relative to three underlying system parameters: traffic flow, communication radio range, and penetration ratio of participating vehicles. Lastly, the enhanced GATIS-V2V model is compared with the centralized traffic information system. This research found that the enhanced GATIS-V2V model outperforms the basic model in terms of travel time savings and produces more consistent and robust system output under non-recurrent traffic states (i.e., traffic incident) in the simple traffic network. This research also identified that the traffic incident detection time and driver's route choice rule are the most crucial factors influencing the system performance. As expected, as traffic flow and penetration ratio increase, the system becomes more efficient, with non-participating vehicles also benefiting from the re-routing of participating vehicles. The communication radio ranges considered were found not to significantly influence system operations in the studied traffic network. Finally, it is found that the decentralized GATIS-V2V model has similar performance to the centralized model even under low flow, short radio range, and low penetration ratio cases. This implies that a dynamic infrastructure-based traffic information system could replace a fixed infrastructure-based traffic information system, allowing for considerable savings in fixed costs and ready expansion of the system off of the main network corridors.

This thesis descries the evolution and testing of a fully portable, inductive loop vehicle counter system. As a component of the NFS Embedded Distributed Simulation for Transportation System Management project, the system's cellular modem transmits real-time data to servers at Georgia Institute of Technology. From there, the data can be fed into simulations predicting travel behavior. Researchers revised both the detector circuit, and the temporary, reusable loop pad several times over multiple rounds of field testing. The final tested version of this system demonstrates the efficacy of uncommonly small inductive loops. When paired with a reliable data transmission channel, the system was shown to capture nearly 96% of actual through traffic.