Esta dissertação apresenta uma análise sobre a modelagem de travessias de pedestres. Os simuladores de tráfego recentemente incorporaram modelos de pedestres, porém, as habilidades e limitações para a representação dos comportamentos em travessias não foram analisadas e apresentadas na literatura. Esta dissertação é composta por três artigos, nos quais são apresentados: (i) caracterizações dos comportamentos de pedestres e motoristas em travessias, (ii) análises dos modelos que representam o comportamento em travessias, (iii) a metodologia desenvolvida para coleta de dados de travessias, (iv) a calibração dos modelos de movimentação de pedestres, aceitação de brechas e da decisão dos pedestres sobre desrespeitar semáforos em travessias semaforizadas de pedestres e (v) uma análise da habilidade e limitações dos modelos para representar os atrasos de pedestres em travessias semaforizadas. Observou-se que a representação dos comportamentos em travessias depende de diversos modelos, que representam a escolha de rotas dos pedestres, movimentação, aceitação de brechas, decisão dos motoristas sobre dar preferência em travessias e decisão dos pedestres sobre obedecer aos semáforos. Os modelos de aceitação de brechas de pedestres utilizados, por exemplo, são baseados nos modelos desenvolvidos para representar interseções não semaforizadas de veículos e simplificam consideravelmente os comportamentos de pedestres e motoristas. As dificuldades para coletar dados de pedestres em travessias é um dos fatores que limitam o desenvolvimento de novos modelos. Nesta dissertação foi desenvolvida uma metodologia semiautomática, na qual os pedestres e veículos são rastreados com softwares livres, que facilita a obtenção de dados. As coletas de dados desenvolvidas permitiram a calibração dos modelos do VISSIM para representar uma travessia semaforizada com múltiplas faixas de tráfego, onde a maioria dos pedestres não obedece ao semáforo. Apesar de que parâmetros como as velocidades desejadas e brechas mínimas aceitas pelos pedestres predefinidas nos modelos foram inadequadas para a simulação da travessia, a calibração da movimentação, aceitação de brechas e decisão sobre desrespeitar os semáforos permitiu que o atraso médio de pedestres fosse representado adequadamente. As principais limitações dos modelos estão associadas à representação do comportamento oportunista, como no caso dos pedestres que desrespeitam os semáforos, analisados nesta dissertação. / This dissertation presents an analisys about pedestrians’ crossings modeling. Traffic simulators recently incorporated pedestrians’ models, however, the hability and limitations to represent conflicts in pedestrians’ crossings have not been analysed and presented on the literature. This dissertation is composed by three articles that present: (i) characterizations of pedestrians as motorist behaviors on crossings, (ii) analisys of models that represent the behavior on crossings, (iii) a methodology developed for collecting data from crossings, (iv) a calibration of pedestrian movement, gap acceptance and the decision about disrespecting siganilzation on signalized pedestrian crossings and (v) an analysis of the hability and limitations of the models in representing pedestrian’s delays. It is observed that the representations of the behaviors on pedestrian crossings depend on several models that represent the route choice, movement, gap acceptance, motorist yielding and the decision os pedestrians about obeying the signalization. The models used on simulation correspond, generally, for deterministic models that simplify pedestrians and motorists’ behavior. Pedestrians’ gap acceptance models, for example, were based on models developed to represent vehicle’s unsignilized intersections. Difficulties in pedestrians’ data collection are one factor that limits the development of new models. On this dissertation, a semi-automatic methodology was developed, on which pedestrians and vehicles are tracked with free software that facilitates data obtainment. The collected data allowed the calibration of VISSIM models to represent a signalized pedestrian crossing with multiple traffic lanes where most pedestrians do not obey the siganization. Altough default parameters as desired speed and minimum rear gap accepted by pedestrians were inadequate for the crossing simulation, the calibration of movement, gap acceptance and the decision about disrespecting the signalization allowed the representation of adequate average delays. The main limitations of the models are associated to the representation of opportunistic behavior, as pedestrians disrespecting sigalizations, analyzed on this dissertation.

Esta dissertação apresenta uma análise sobre a modelagem de travessias de pedestres. Os simuladores de tráfego recentemente incorporaram modelos de pedestres, porém, as habilidades e limitações para a representação dos comportamentos em travessias não foram analisadas e apresentadas na literatura. Esta dissertação é composta por três artigos, nos quais são apresentados: (i) caracterizações dos comportamentos de pedestres e motoristas em travessias, (ii) análises dos modelos que representam o comportamento em travessias, (iii) a metodologia desenvolvida para coleta de dados de travessias, (iv) a calibração dos modelos de movimentação de pedestres, aceitação de brechas e da decisão dos pedestres sobre desrespeitar semáforos em travessias semaforizadas de pedestres e (v) uma análise da habilidade e limitações dos modelos para representar os atrasos de pedestres em travessias semaforizadas. Observou-se que a representação dos comportamentos em travessias depende de diversos modelos, que representam a escolha de rotas dos pedestres, movimentação, aceitação de brechas, decisão dos motoristas sobre dar preferência em travessias e decisão dos pedestres sobre obedecer aos semáforos. Os modelos de aceitação de brechas de pedestres utilizados, por exemplo, são baseados nos modelos desenvolvidos para representar interseções não semaforizadas de veículos e simplificam consideravelmente os comportamentos de pedestres e motoristas. As dificuldades para coletar dados de pedestres em travessias é um dos fatores que limitam o desenvolvimento de novos modelos. Nesta dissertação foi desenvolvida uma metodologia semiautomática, na qual os pedestres e veículos são rastreados com softwares livres, que facilita a obtenção de dados. As coletas de dados desenvolvidas permitiram a calibração dos modelos do VISSIM para representar uma travessia semaforizada com múltiplas faixas de tráfego, onde a maioria dos pedestres não obedece ao semáforo. Apesar de que parâmetros como as velocidades desejadas e brechas mínimas aceitas pelos pedestres predefinidas nos modelos foram inadequadas para a simulação da travessia, a calibração da movimentação, aceitação de brechas e decisão sobre desrespeitar os semáforos permitiu que o atraso médio de pedestres fosse representado adequadamente. As principais limitações dos modelos estão associadas à representação do comportamento oportunista, como no caso dos pedestres que desrespeitam os semáforos, analisados nesta dissertação. / This dissertation presents an analisys about pedestrians’ crossings modeling. Traffic simulators recently incorporated pedestrians’ models, however, the hability and limitations to represent conflicts in pedestrians’ crossings have not been analysed and presented on the literature. This dissertation is composed by three articles that present: (i) characterizations of pedestrians as motorist behaviors on crossings, (ii) analisys of models that represent the behavior on crossings, (iii) a methodology developed for collecting data from crossings, (iv) a calibration of pedestrian movement, gap acceptance and the decision about disrespecting siganilzation on signalized pedestrian crossings and (v) an analysis of the hability and limitations of the models in representing pedestrian’s delays. It is observed that the representations of the behaviors on pedestrian crossings depend on several models that represent the route choice, movement, gap acceptance, motorist yielding and the decision os pedestrians about obeying the signalization. The models used on simulation correspond, generally, for deterministic models that simplify pedestrians and motorists’ behavior. Pedestrians’ gap acceptance models, for example, were based on models developed to represent vehicle’s unsignilized intersections. Difficulties in pedestrians’ data collection are one factor that limits the development of new models. On this dissertation, a semi-automatic methodology was developed, on which pedestrians and vehicles are tracked with free software that facilitates data obtainment. The collected data allowed the calibration of VISSIM models to represent a signalized pedestrian crossing with multiple traffic lanes where most pedestrians do not obey the siganization. Altough default parameters as desired speed and minimum rear gap accepted by pedestrians were inadequate for the crossing simulation, the calibration of movement, gap acceptance and the decision about disrespecting the signalization allowed the representation of adequate average delays. The main limitations of the models are associated to the representation of opportunistic behavior, as pedestrians disrespecting sigalizations, analyzed on this dissertation.

Esta dissertação apresenta uma análise sobre a modelagem de travessias de pedestres. Os simuladores de tráfego recentemente incorporaram modelos de pedestres, porém, as habilidades e limitações para a representação dos comportamentos em travessias não foram analisadas e apresentadas na literatura. Esta dissertação é composta por três artigos, nos quais são apresentados: (i) caracterizações dos comportamentos de pedestres e motoristas em travessias, (ii) análises dos modelos que representam o comportamento em travessias, (iii) a metodologia desenvolvida para coleta de dados de travessias, (iv) a calibração dos modelos de movimentação de pedestres, aceitação de brechas e da decisão dos pedestres sobre desrespeitar semáforos em travessias semaforizadas de pedestres e (v) uma análise da habilidade e limitações dos modelos para representar os atrasos de pedestres em travessias semaforizadas. Observou-se que a representação dos comportamentos em travessias depende de diversos modelos, que representam a escolha de rotas dos pedestres, movimentação, aceitação de brechas, decisão dos motoristas sobre dar preferência em travessias e decisão dos pedestres sobre obedecer aos semáforos. Os modelos de aceitação de brechas de pedestres utilizados, por exemplo, são baseados nos modelos desenvolvidos para representar interseções não semaforizadas de veículos e simplificam consideravelmente os comportamentos de pedestres e motoristas. As dificuldades para coletar dados de pedestres em travessias é um dos fatores que limitam o desenvolvimento de novos modelos. Nesta dissertação foi desenvolvida uma metodologia semiautomática, na qual os pedestres e veículos são rastreados com softwares livres, que facilita a obtenção de dados. As coletas de dados desenvolvidas permitiram a calibração dos modelos do VISSIM para representar uma travessia semaforizada com múltiplas faixas de tráfego, onde a maioria dos pedestres não obedece ao semáforo. Apesar de que parâmetros como as velocidades desejadas e brechas mínimas aceitas pelos pedestres predefinidas nos modelos foram inadequadas para a simulação da travessia, a calibração da movimentação, aceitação de brechas e decisão sobre desrespeitar os semáforos permitiu que o atraso médio de pedestres fosse representado adequadamente. As principais limitações dos modelos estão associadas à representação do comportamento oportunista, como no caso dos pedestres que desrespeitam os semáforos, analisados nesta dissertação. / This dissertation presents an analisys about pedestrians’ crossings modeling. Traffic simulators recently incorporated pedestrians’ models, however, the hability and limitations to represent conflicts in pedestrians’ crossings have not been analysed and presented on the literature. This dissertation is composed by three articles that present: (i) characterizations of pedestrians as motorist behaviors on crossings, (ii) analisys of models that represent the behavior on crossings, (iii) a methodology developed for collecting data from crossings, (iv) a calibration of pedestrian movement, gap acceptance and the decision about disrespecting siganilzation on signalized pedestrian crossings and (v) an analysis of the hability and limitations of the models in representing pedestrian’s delays. It is observed that the representations of the behaviors on pedestrian crossings depend on several models that represent the route choice, movement, gap acceptance, motorist yielding and the decision os pedestrians about obeying the signalization. The models used on simulation correspond, generally, for deterministic models that simplify pedestrians and motorists’ behavior. Pedestrians’ gap acceptance models, for example, were based on models developed to represent vehicle’s unsignilized intersections. Difficulties in pedestrians’ data collection are one factor that limits the development of new models. On this dissertation, a semi-automatic methodology was developed, on which pedestrians and vehicles are tracked with free software that facilitates data obtainment. The collected data allowed the calibration of VISSIM models to represent a signalized pedestrian crossing with multiple traffic lanes where most pedestrians do not obey the siganization. Altough default parameters as desired speed and minimum rear gap accepted by pedestrians were inadequate for the crossing simulation, the calibration of movement, gap acceptance and the decision about disrespecting the signalization allowed the representation of adequate average delays. The main limitations of the models are associated to the representation of opportunistic behavior, as pedestrians disrespecting sigalizations, analyzed on this dissertation.

Indiana University-Purdue University Indianapolis (IUPUI) / For pedestrians and autonomous vehicles (AVs) to co-exist harmoniously and safely in the real-world, AVs will need to not only react to pedestrian actions, but also anticipate their intentions. In this thesis, we propose to use rich visual and pedestrian-environment interaction features to improve pedestrian crossing intention prediction from the ego-view.We do so by combining visual feature extraction, graph modeling of scene objects and their relationships, and feature encoding as comprehensive inputs for an LSTM encoder-decoder network. Pedestrians react and make decisions based on their surrounding environment, and the behaviors of other road users around them. The human-human social relationship has al-ready been explored for pedestrian trajectory prediction from the bird’s eye view in stationary cameras. However, context and pedestrian-environment relationships are often missing incurrent research into pedestrian trajectory, and intention prediction from the ego-view. To map the pedestrian’s relationship to its surrounding objects we use a star graph with the pedestrian in the center connected to all other road objects/agents in the scene. The pedestrian and road objects/agents are represented in the graph through visual features extracted using state of the art deep learning algorithms. We use graph convolutional networks, and graph autoencoders to encode the star graphs in a lower dimension. Using the graph en-codings, pedestrian bounding boxes, and human pose estimation, we propose a novel model that predicts pedestrian crossing intention using not only the pedestrian’s action behaviors(bounding box and pose estimation), but also their relationship to their environment. Through tuning hyperparameters, and experimenting with different graph convolutions for our graph autoencoder, we are able to improve on the state of the art results. Our context-driven method is able to outperform current state of the art results on benchmark datasetPedestrian Intention Estimation (PIE). The state of the art is able to predict pedestrian crossing intention with a balanced accuracy (to account for dataset imbalance) score of 0.61, while our best performing model has a balanced accuracy score of 0.79. Our model especially outperforms in no crossing intention scenarios with an F1 score of 0.56 compared to the state of the art’s score of 0.36. Additionally, we also experiment with training the state of the art model and our model to predict pedestrian crossing action, and intention jointly. While jointly predicting crossing action does not help improve crossing intention prediction, it is an important distinction to make between predicting crossing action versus intention.

Approximately 47,700 pedestrians were killed between the years of 2000 - 2009. School buses are one of the safest modes of transportation (National Highway Traffic Safety Administration, 2004). However, the Central Florida school district eliminated bus transportation within the 2-mile radius from schools just last year. Children must prepare for an alternative mode of transportation; walking and biking. The purpose of this research was two-fold. First to develop an online safety training program for elementary school children; and second, a self-report questionnaire was constructed and piloted to measure how safety training and school infrastructure affects students' pedestrian risk-taking attitudes and risk perceptions to avoid the dangers of walking and biking to and from school. A 2x2 Factorial Multivariate Analysis of Variance (MANOVA) was used to test two categorical independent variables (safety awareness training, school infrastructure) for each of the two continuous dependent variables (pedestrian risk-taking attitudes and risk perceptions of pedestrian behavior). Using data from the pilot study, the researcher developed, self-reported questionnaires demonstrated that there was a significant difference between schools. Those receiving the training had lower mean scores in risk-taking attitudes than those who did not receive the training. Regardless of intervention, School 2 (complete infrastructure) takes fewer risks than School 1(incomplete infrastructure). The mean difference between groups was not statistically significant.

No description available.

Thesis (M. Arch.)--University of Hong Kong, 2001. / Includes bibliographical references.

published_or_final_version / Transport Policy and Planning / Master / Master of Arts in Transport Policy and Planning

published_or_final_version / Transport Policy and Planning / Master / Master of Arts in Transport Policy and Planning

Thesis (M.Arch.)--University of Hong Kong, 2001. / Includes bibliographical references. Also available in print.