With the advent of Advanced Traveler Information Systems (ATIS), short-term travel time prediction is becoming increasingly important. Travel time can be obtained directly from instrumented test vehicles, license plate matching, probe vehicles etc., or from indirect methods such as loop detectors. Because of their wide spread deployment, travel time estimation from loop detector data is one of the most widely used methods. However, the major criticism about loop detector data is the high probability of error due to the prevalence of equipment malfunctions. This dissertation presents methodologies for estimating and predicting travel time from the loop detector data after correcting for errors. The methodology is a multi-stage process, and includes the correction of data, estimation of travel time and prediction of travel time, and each stage involves the judicious use of suitable techniques. The various techniques selected for each of these stages are detailed below. The test sites are from the freeways in San Antonio, Texas, which are equipped with dual inductance loop detectors and AVI. ?? Constrained non-linear optimization approach by Generalized Reduced Gradient (GRG) method for data reduction and quality control, which included a check for the accuracy of data from a series of detectors for conservation of vehicles, in addition to the commonly adopted checks. ?? A theoretical model based on traffic flow theory for travel time estimation for both off-peak and peak traffic conditions using flow, occupancy and speed values obtained from detectors. ?? Application of a recently developed technique called Support Vector Machines (SVM) for travel time prediction. An Artificial Neural Network (ANN) method is also developed for comparison. Thus, a complete system for the estimation and prediction of travel time from loop detector data is detailed in this dissertation. Simulated data from CORSIM simulation software is used for the validation of the results.

The purpose of this research is to develop methodologies for applying traffic flow theories to various ITS categories through the utilization of evolving surveillance technologies. This integration of theory, measurement and application has been overlooked since the advent of ITS because of the number of disciplines involved. In this context, the following illustrative methodologies are selected, developed and presented in this study: - a methodology for automatic measurement of major spatial traffic variables for the present and the future implementation of various ITS functional areas, in general; and - a methodology for real-time link and incident specific freeway diversion in conjunction with freeway incident management, in particular. The first methodology includes the development of a dynamic flow model based on stochastic queuing theory and the principle of conservation of vehicles. An inductive modeling approach adapted here utilizes geometric interpretations of cumulative arrival-departure diagrams which have been drawn directly from surveillance data. The advantages of this model are real-time applicability and transportability as well as ease of use. Analysis results show that the estimates are in qualitative and quantitative agreement with the empirical data measured at 30-second intervals. The analytical expression for link travel times satisfies traffic dynamics where the new form of the equation of conservation of vehicles has been derived. This methodology has potential applicable to automatic traffic control and automatic incident detection. The methodology is then applied to freeway diversion in real-time in conjunction with freeway incident management. The proposed new form of the equation of conservation of vehicles is applied to detect recurring or non-recurring congestion analytically. The principle of conservation of vehicles is applied to develop the concept of progression and retrogression of incident domain, which turns out to be compatible with traditional shock wave traffic mechanism during incidents. The link and incident specific diversion methodology is achieved by using a delay diagram and volume-travel time curves, which can be plotted per link per incident. The use of such graphic aids makes problem solving much easier and clearer. The dynamic traffic flow model developed here can also be applied to estimate travel times during incidents as a function of time. The development of a computer program for freeway diversion concludes this research. / Ph. D.

The use of GPS probes in traffic management is growing rapidly as the required data collection infrastructure is increasingly in place, with significant number of mobile sensors moving around covering expansive areas of the road network. Many travelers carry with them at least one device with a built-in GPS receiver. Furthermore, vehicles are becoming more and more location aware. Vehicles in commercial fleets are now routinely equipped with GPS. Travel time is important information for various actors of a transport system, ranging from city planning, to day to day traffic management, to individual travelers. They all make decisions based on average travel time or variability of travel time among other factors. AVI (Automatic Vehicle Identification) systems have been commonly used for collecting point-to-point travel time data. Floating car data (FCD) -timestamped locations of moving vehicles- have shown potential for travel time estimation. Some advantages of FCD compared to stationary AVI systems are that they have no single point of failure and they have better network coverage. Furthermore, the availability of opportunistic sensors, such as GPS, makes the data collection infrastructure relatively convenient to deploy. Currently, systems that collect FCD are designed to transmit data in a limited form and relatively infrequently due to the cost of data transmission. Thus, reported locations are far apart in time and space, for example with 2 minutes gaps. For sparse FCD to be useful for transport applications, it is required that the corresponding probes be matched to the underlying digital road network. Matching such data to the network is challenging. This thesis makes the following contributions: (i) a map-matching and path inference algorithm, (ii) a method for route travel time estimation, (iii) a fixed point approach for joint path inference and travel time estimation, and (iv) a method for fusion of FCD with data from automatic number plate recognition. In all methods, scalability and overall computational efficiency are considered among design requirements. Throughout the thesis, the methods are used to process FCD from 1500 taxis in Stockholm City. Prior to this work, the data had been ignored because of its low frequency and minimal information. The proposed methods proved that the data can be processed and transformed into useful traffic information. Finally, the thesis implements the main components of an experimental ITS laboratory, called iMobility Lab. It is designed to explore GPS and other emerging data sources for traffic monitoring and control. Processes are developed to be computationally efficient, scalable, and to support real time applications with large data sets through a proposed distributed implementation. / <p>QC 20150525</p>

This dissertation is concerned with the problem of estimating travel time per links in urban context using sparsely sampled GPS data. One of the challenges in this thesis is use the sparsely sampled data. A part of this research work, i developed a digital map with its new geographic information system (GIS), dealing with map-matching problem, where we come out with an enhancement tecnique, and also the shortest path problem.The thesis research work was conduct within the project PUMAS, which is an avantage for our research regarding the collection process of our data from the real world field and also in making our tests. The project PUMAS (Plate-forme Urbaine de Mobilité Avancée et Soutenable / Urban Platform for Sustainable and Advanced Mobility) is a preindustrial project that has the objective to inform about the traffic situation and also to develop an implement a platform for sustainable mobility in order to evaluate it in the region, specifically Rouen, France. The result is a framework for any traffic controller or manager and also estimation researcher to access vast stores of data about the traffic estimation, forecasting and status.

[pt] Ônibus, equipados com dispositivos GPS ativos que transmitem continuamente a sua posição, podem ser entendidos como sensores móveis de trânsito. De fato, as trajetórias dos ônibus fornecem uma fonte de dados útil para analisar o trânsito na rede de ônibus de uma cidade, dado que as autoridades de trânsito da cidade disponibilizem as trajetórias de forma aberta, oportuna e contínua. Neste contexto, esta tese propõe uma abordagem que usa os dados de GPS dos ônibus para analisar e monitorar a rede de ônibus de uma cidade. Ela combina algoritmos de grafos, técnicas de mineração de dados geoespaciais e métodos estatísticos. A principal contribuição desta tese é uma definição detalhada de operações e algoritmos para analisar e monitorar o tráfego na rede de ônibus, especificamente: (1) modelagem, análise e segmentaçãoda rede de ônibus; (2) mineração do conjunto de dados de trajetória de ônibus para descobrir padrões de tráfego; (3) detecção de anomalias de trânsito, classificação de acordo com sua gravidade, e avaliação do seu impacto; (4) manutenção e comparação de diferentes versões da rede de ônibus e dos seus padrões de tráfego para ajudar os planejadores urbanos a avaliar as mudanças. Uma segunda contribuição é a descrição de experimentos realizados para a rede de ônibus da Cidade do Rio de Janeiro, utilizando trajetórias de ônibus correspondentes ao período de junho de 2014 até fevereiro de 2017, disponibilizadas pela Prefeitura do Rio de Janeiro. Os resultados obtidos corroboram a utilidade da abordagem proposta para analisar e monitorar a rede de ônibus de uma cidade, o que pode ajudar os gestores do trânsito e as autoridades municipais a melhorar os planos de controle de trânsito e de mobilidade urbana. / [en] Buses, equipped with active GPS devices that continuously transmit their position, can be understood as mobile traffic sensors. Indeed, bus trajectories provide a useful data source for analyzing traffic in the bus network of a city, if the city traffic authority makes the bus trajectories available openly, timely and in a continuous way. In this context, this thesis proposes a bus GPS data-driven approach for analyzing and monitoring the bus network of a city. It combines graph algorithms, geospatial data mining techniques and statistical methods. The major contribution of this thesis is a detailed discussion of key operations and algorithms for modeling, analyzing and monitoring bus network traffic, specifically: (1) modelling, analyzing, and segmentation of the bus network; (2) mining the bus trajectory dataset to uncover traffic patterns; (3) detecting traffic anomalies, classifying them according to their severity, and estimating their impact; (4) maintaining and comparing different versions of the bus network and traffic patterns to help urban planners assess changes. Another contribution is the description of experiments conducted for the bus network of the City of Rio de Janeiro, using bus trajectories obtained from June 2014 to February 2017, which have been made available by the City Hall of Rio de Janeiro. The results obtained corroborate the usefulness of the proposed approach for analyzing and monitoring the bus network of a city, which may help traffic managers and city authorities improve traffic control and urban mobility plans.

No description available.

Cette thèse porte sur le problème de l'estimation des temps de parcours, de véhicules, par section de route dans un contexte urbain, en utilisant les données GPS à faible densité d’échantillon. L'un des défis de cette thèse est d'utiliser ce genre de données. Dans le cadre de ce travail de recherche, j'ai développé une carte numérique avec son nouveau système d'information géographique (SIG), qui traite la problématique du map-matching, où nous avons apporté des améliorations, ainsi que le problème du plus court chemin.La thèse s'inscrit dans le cadre du projet PUMAS (Plate-forme Urbaine de Mobilité Avancée et Soutenable), ce qui est un avantage pour nos recherches en ce qui concerne le processus de collecte de données réelles sur le terrain ainsi que pour faire nos tests. Le projet PUMAS est un projet préindustriel qui a pour objectif d'informer sur la situation du trafic mais également de développer et de mettre en œuvre une plate-forme de mobilité durable afin de l'évaluer dans la région, notamment à Rouen, France. Le résultat offre un cadre pour tout contrôleur de la situation, gestionnaire ou chercheur pour accéder à de vastes réserves de données sur l'estimation du flux du trafic, sur les prévisions et sur l'état du trafic. / This dissertation is concerned with the problem of estimating travel time per links in urban context using sparsely sampled GPS data. One of the challenges in this thesis is use the sparsely sampled data. A part of this research work, i developed a digital map with its new geographic information system (GIS), dealing with map-matching problem, where we come out with an enhancement tecnique, and also the shortest path problem.The thesis research work was conduct within the project PUMAS, which is an avantage for our research regarding the collection process of our data from the real world field and also in making our tests. The project PUMAS (Plate-forme Urbaine de Mobilité Avancée et Soutenable / Urban Platform for Sustainable and Advanced Mobility) is a preindustrial project that has the objective to inform about the traffic situation and also to develop an implement a platform for sustainable mobility in order to evaluate it in the region, specifically Rouen, France. The result is a framework for any traffic controller or manager and also estimation researcher to access vast stores of data about the traffic estimation, forecasting and status.

Cette thèse se penche sur les modèles de choix, des distributions sur des ensembles d'alternatives. Les modèles de choix sur les processus décisionnels de Markov (MDP) peuvent décomposer de très grands espaces alternatifs en procédures étape par étape conçues pour non seulement combattre la malédiction de la dimensionnalité mais aussi pour mieux refléter la dynamique sous-jacente. La première partie est consacrée à l'estimation du temps de trajet dans le cadre de la modélisation du choix de chemin. Les modèles de choix de chemin sont des modèles de choix sur l'ensemble des chemins utilisés pour modéliser le flux de circulation. Intuitivement, le temps de trajet est l'une des caractéristiques les plus importantes lors du choix des chemins, mais les temps de trajet ne sont pas toujours connus. En revanche, le cadre classique suppose que ces deux étapes sont séquentielles, car les temps de trajet des arcs font partie de l'entrée du processus d'estimation du choix de chemin. Pourtant, les interdépendances complexes signifient que ce modèle de choix de chemin peut complémenter toute observation lors de l'estimation des temps de trajet. Nous construisons un modèle statistique pour l'estimation du temps de trajet et proposons de marginaliser les caractéristiques non observées. En utilisant ces idées, nous montrons que nous sommes capables d'apprendre des modèles de choix de chemin sans observer de chemins réels et à différentes granularités. La deuxième partie se concentre sur les échecs des MDP régularisés et comment la régularisation peut avoir des effets secondaires inattendus, tels que la divergence dans les chemins stochastiques les plus courts ou des fonctions de valeur déraisonnablement grandes. Les MDP régularisés ne sont rien d'autre qu'une application des modèles de choix aux MDP. Ils sont utilisés dans l'apprentissage par renforcement (RL) pour obtenir, entre autres choses, un modèle de choix sur les trajectoires possibles pour l'apprentissage par renforcement inverse, transférer des connaissances préalables au modèle, ou obtenir des politiques qui exploitent tous les objectifs dans l'environnement. Ces effets secondaires sont exacerbés dans les espaces d'action dépendants de l'état. Comme mesure d'atténuation, nous introduisons deux transformations potentielles, et nous évaluons leur performance sur un problème de conception de médicaments. / This thesis delves on choice models, distributions on sets of alternatives. Choice models on Markov decision processes (MDPs) can break down very large alternative spaces into step-by-step procedures designed to not only tackle the curse of dimensionality but also to reflect the underlying dynamics better. The first part is devoted to travel time estimation as part of path choice modeling. Path choice models are choice models on the set of paths used to model traffic flow. Intuitively, travel time is one of the more important features when choosing paths, yet travel times are not always known. In contrast, the classical setting assumes that these two steps are sequential, as arc travel times are part of the input of the path choice estimation process. Yet the intricate interdependences mean that that path choice model can complement any observation when estimating travel times. We build a statistical model for travel time estimation and propose marginalizing the unobserved features. Using these ideas, we show that we are able to learn path choice models without observing actual paths and at different granularity. The second part focuses on the failings of regularized MDPs and how regularization may have unexpected side effects, such as divergence in stochastic shortest paths or unreasonably large value functions. Regularized MDPs are nothing but an application of choice models to MDPs. They are used in reinforcement learning (RL) to get, among other things, a choice model on possible trajectories for inverse reinforcement learning, transfer prior knowledge to the model, or to get policies that exploit all goals in the environment. These side effects are exacerbated in state-dependent action spaces. As a mitigation, we introduce two potential transformations, and we benchmark their performance on a drug design problem.