This thesis proposes a novel multiple traffic light recognition system based on videos captured by a monocular camera. Advanced driver assistance system (ADAS) and autonomous driving system (ADS) are becoming increasingly important to help drivers maneuvering vehicles and increase the vehicle and road safety in modern life. Traffic light recognition system is a significant part of ADAS and ADS, which can detect traffic light on the road and recognize different types of traffic lights to provide useful signal information for drivers. The proposed method can be applied to real complex environment only based on a monocular camera and is tested in real-world scenarios. This system consists of three parts: multiple traffic light detection, multi-target tracking and state classification. For the first step, a supervised machine learning method, support vector machine (SVM) with two integral features - histogram of oriented gradients (HOG) and histogram of CIELAB color space (HCIELAB), are used to detect traffic lights in the captured image. Then, a new multi-target tracking algorithm is presented to improve the accuracy of detection, reduce the number of false alarm and missing targets, by means of nearest neighbor data association, motion model analysis and Lucas-Kanade optical flow tracking and the region of interest (ROI) prediction. Finally, a SVM-based and a convolution neural network (CNN) based classifiers are introduced to classify the state of traffic lights, that provides the stop, go, warning, straight and turn information. Various experiments have been conducted to demonstrate the practicability of the proposed method. Both GPU-based and CPU-based programming can run real-time on the real street environment. / Thesis / Master of Applied Science (MASc)

As the number of vehicles increases rapidly, traffic congestion has become a serious problem in a city. Over the past years, a considerable number of studies have been made on traffic light setting. The traffic light setting problem is to investigate how to set the given traffic lights such that the total waiting time of vehicles on the roads is minimized. In this thesis, we use a graph model to represent the traffic network. On this model, some characteristics of the setting problem can be presented and analyzed. We first devise a branch and bound algorithm for obtaining the optimal solution of the traffic light setting problem. In addition, the genetic algorithm (GA), the particle swarm optimization (PSO) and the ant colony optimization (ACO) algorithm are also adopted to get the near optimal solution. Then, to extend this model, we add the assumption that each vehicle can change its direction. By comparing the results of various algorithms, we can study the impact of these algorithms on the traffic light setting problem. In our experiments, we also transform the map of Kaohsiung city into our graph model and test each algorithm on this graph.

Training a large dataset has become a serious issue for researchers because it requires large memories and can take a long time for computing. People are trying to process large scale dataset not only by changing programming model, such as using MapReduce and Hadoop, but also by designing new algorithms that can retain performance with less complexity and runtime. In this thesis, we present implementations of incremental learning and online learning methods to classify a large traffic light dataset for traffic light recognition. The introduction part includes the concepts and related works of incremental learning and online learning. The main algorithm is a modification of IMORL incremental learning model to enhance its performance over the learning process of our application. Then we briefly discuss how the traffic light recognition algorithm works and the problem we encounter during training. Rather than focusing on incremental learning, which uses batch to batch data during training procedure, we introduce Pegasos, an online style primal gradient-based support vector machine method. The performance of Pegasos for classification is extraordinary and the number of instances it uses for training is relatively small. Therefore, Pegasos is the recommended solution to the large dataset training problem.

In large urbanized cities, a major problem that affects the economy and health of all citizens is vehicular congestion. This is because the traffic light cycles are not adequate. In the present study, we seek to optimize traffic light cycles based on simulation, in order to improve vehicle flow. For this, the PTV Vissim 9.0 software was used as a simulator and the Synchro 10.0 software to determine the initial optimal traffic light cycle. Through several runs and having as variables the length of queues, delay times and the average speed, the optimal traffic light cycle could be found for the study area. The results obtained reflect a 14% reduction in delay times and 10% in queue lengths. On the other hand, the average vehicle speed increased by 10.56%. All this represents an improvement in the service level of the study intersections.

Este artículo presenta la evaluación del tráfico vehicular en dos intersecciones semaforizadas ubicadas en la ciudad de Lima; una de las principales ciudades del mundo con gran congestión vehicular; mediante un modelo microscópico desarrollado con el software Vissim 9.0. La finalidad de esta investigación es proponer la implementación de un sistema de semaforización inteligente para mejorar los niveles de servicio y disminuir la congestión vehicular. El proceso de construcción del modelo consiste en cuatro etapas. Primero, es el trabajo previo, en la cual se realizó la recolección de datos de campo. Luego, se construyó el modelo mediante el software Vissim 9.0. Después, se realizó la calibración y validación del modelo en ambas intersecciones en base a parámetros psicofísicos para certificar que los resultados del modelo se asemejen a las condiciones reales de las intersecciones. Por último, se incorporaron los nuevos flujos futuros en el modelo cada 15 minutos los cuales se simularon y evaluaron con los NS obtenidos, que incorporan la propuesta de mejora. Esencialmente los niveles de servicio de las intersecciones mejoran, determinando cada 15 minutos durante la hora pico nuevos ciclos semaforizados optimizados con el programa Synchro 10.0 aplicados a Vissim 9.0, pasando de un NS E a D, esto se deduce a partir de la disminución de longitudes de colas en todos los accesos y la reducción de las demoras de viaje, las cuales fueron 14% en la Av. Javier Prado – Ca. Las Flores y 24% en la Av. Javier Prado – Ca. Las Palmeras. / This paper presents the evaluation of vehicular traffic on two signalized intersections located in the city of Lima; one of the world's major cities with high traffic congestion; by microscopic model developed with Vissim 9.0 software. The purpose of this research is to propose the implementation of an intelligent traffic signals to improve service levels and reduce traffic congestion. The process of building the model consists of four stages. First, it is the previous work, in which the field data collection was performed. The model was then constructed by Vissim 9.0 software. After, calibration and validation of the model was performed on both intersections based on psychophysical parameters to certify that the model results resemble the actual conditions of intersections. Finally, new future flows were incorporated into the model every 15 minutes which were simulated and evaluated with the NS obtained, incorporating the proposed improvement. Essentially service levels intersections improve, determining every 15 minutes during rush hour new signalized cycles optimized with Synchro 10.0 program applied to Vissim 9.0, from an NS E to D, It follows from decreasing lengths queues at all entrances and reducing travel delays, which were 14% at Av. Javier Prado -. Ca. Las Flores and 24% on Av. Javier Prado -. Ca. Las Palmeras. / Trabajo de investigación

Este artículo presenta la evaluación del tráfico vehicular en dos intersecciones semaforizadas ubicadas en la ciudad de Lima; una de las principales ciudades del mundo con gran congestión vehicular; mediante un modelo microscópico desarrollado con el software Vissim 9.0. La finalidad de esta investigación es proponer la implementación de un sistema de semaforización inteligente para mejorar los niveles de servicio y disminuir la congestión vehicular. El proceso de construcción del modelo consiste en cuatro etapas. Primero, es el trabajo previo, en la cual se realizó la recolección de datos de campo. Luego, se construyó el modelo mediante el software Vissim 9.0. Después, se realizó la calibración y validación del modelo en ambas intersecciones en base a parámetros psicofísicos para certificar que los resultados del modelo se asemejen a las condiciones reales de las intersecciones. Por último, se incorporaron los nuevos flujos futuros en el modelo cada 15 minutos los cuales se simularon y evaluaron con los NS obtenidos, que incorporan la propuesta de mejora. Esencialmente los niveles de servicio de las intersecciones mejoran, determinando cada 15 minutos durante la hora pico nuevos ciclos semaforizados optimizados con el programa Synchro 10.0 aplicados a Vissim 9.0, pasando de un NS E a D, esto se deduce a partir de la disminución de longitudes de colas en todos los accesos y la reducción de las demoras de viaje, las cuales fueron 14% en la Av. Javier Prado – Ca. Las Flores y 24% en la Av. Javier Prado – Ca. Las Palmeras. / This paper presents the evaluation of vehicular traffic on two signalized intersections located in the city of Lima; one of the world's major cities with high traffic congestion; by microscopic model developed with Vissim 9.0 software. The purpose of this research is to propose the implementation of an intelligent traffic signals to improve service levels and reduce traffic congestion. The process of building the model consists of four stages. First, it is the previous work, in which the field data collection was performed. The model was then constructed by Vissim 9.0 software. After, calibration and validation of the model was performed on both intersections based on psychophysical parameters to certify that the model results resemble the actual conditions of intersections. Finally, new future flows were incorporated into the model every 15 minutes which were simulated and evaluated with the NS obtained, incorporating the proposed improvement. Essentially service levels intersections improve, determining every 15 minutes during rush hour new signalized cycles optimized with Synchro 10.0 program applied to Vissim 9.0, from an NS E to D, It follows from decreasing lengths queues at all entrances and reducing travel delays, which were 14% at Av. Javier Prado -. Ca. Las Flores and 24% on Av. Javier Prado -. Ca. Las Palmeras. / Tesis

Este artículo presenta la evaluación del tráfico vehicular en dos intersecciones semaforizadas ubicadas en la ciudad de Lima; una de las principales ciudades del mundo con gran congestión vehicular; mediante un modelo microscópico desarrollado con el software Vissim 9.0. La finalidad de esta investigación es proponer la implementación de un sistema de semaforización inteligente para mejorar los niveles de servicio y disminuir la congestión vehicular. El proceso de construcción del modelo consiste en cuatro etapas. Primero, es el trabajo previo, en la cual se realizó la recolección de datos de campo. Luego, se construyó el modelo mediante el software Vissim 9.0. Después, se realizó la calibración y validación del modelo en ambas intersecciones en base a parámetros psicofísicos para certificar que los resultados del modelo se asemejen a las condiciones reales de las intersecciones. Por último, se incorporaron los nuevos flujos futuros en el modelo cada 15 minutos los cuales se simularon y evaluaron con los NS obtenidos, que incorporan la propuesta de mejora. Esencialmente los niveles de servicio de las intersecciones mejoran, determinando cada 15 minutos durante la hora pico nuevos ciclos semaforizados optimizados con el programa Synchro 10.0 aplicados a Vissim 9.0, pasando de un NS E a D, esto se deduce a partir de la disminución de longitudes de colas en todos los accesos y la reducción de las demoras de viaje, las cuales fueron 14% en la Av. Javier Prado – Ca. Las Flores y 24% en la Av. Javier Prado – Ca. Las Palmeras. / This paper presents the evaluation of vehicular traffic on two signalized intersections located in the city of Lima; one of the world's major cities with high traffic congestion; by microscopic model developed with Vissim 9.0 software. The purpose of this research is to propose the implementation of an intelligent traffic signals to improve service levels and reduce traffic congestion. The process of building the model consists of four stages. First, it is the previous work, in which the field data collection was performed. The model was then constructed by Vissim 9.0 software. After, calibration and validation of the model was performed on both intersections based on psychophysical parameters to certify that the model results resemble the actual conditions of intersections. Finally, new future flows were incorporated into the model every 15 minutes which were simulated and evaluated with the NS obtained, incorporating the proposed improvement. Essentially service levels intersections improve, determining every 15 minutes during rush hour new signalized cycles optimized with Synchro 10.0 program applied to Vissim 9.0, from an NS E to D, It follows from decreasing lengths queues at all entrances and reducing travel delays, which were 14% at Av. Javier Prado -. Ca. Las Flores and 24% on Av. Javier Prado -. Ca. Las Palmeras. / Trabajo de investigación

Traffic congestion has tremendous economic and environmental costs. One way to reduce this congestion is to implement more intelligent traffic light systems. There is significant existing research into different algorithms for controlling traffic lights, but they all use separate systems for performance testing. This paper presents the Rush Hour system, which models a network of roadways and traffic lights as a network of connected routers and endnodes. Several traffic switching algorithms are then tested on the Rush Hour system. As expected, we found that the more intelligent systems were effective at reducing congestion at low and medium levels of traffic. However, they were comparable to more naive algorithms at higher levels of traffic.

abstract: For autonomous vehicles, intelligent autonomous intersection management will be required for safe and efficient operation. In order to achieve safe operation despite uncertainties in vehicle trajectory, intersection management techniques must consider a safety buffer around the vehicles. For truly safe operation, an extra buffer space should be added to account for the network and computational delay caused by communication with the Intersection Manager (IM). However, modeling the worst-case computation and network delay as additional buffer around the vehicle degrades the throughput of the intersection. To avoid this problem, AIM, a popular state-of-the-art IM, adopts a query-based approach in which the vehicle requests to enter at a certain arrival time dictated by its current velocity and distance to the intersection, and the IM replies yes/no. Although this solution does not degrade the position uncertainty, it ultimately results in poor intersection throughput. We present Crossroads, a time-sensitive programming method to program the interface of a vehicle and the IM. Without requiring additional buffer to account for the effect of network and computational delay, Crossroads enables efficient intersection management. Test results on a 1/10 scale model of intersection using TRAXXAS RC cars demonstrates that our Crossroads approach obviates the need for large buffers to accommodate for the network and computation delay, and can reduce the average wait time for the vehicles at a single-lane intersection by 24%. To compare Crossroads with previous approaches, we perform extensive Matlab simulations, and find that Crossroads achieves on average 1.62X higher throughput than a simple VT-IM with extra safety buffer, and 1.36X better than AIM. / Dissertation/Thesis / Masters Thesis Engineering 2017

[pt] Para cruzamentos de veículos isolados, apresenta-se aqui uma lógica de controle dos tempos de passagem das fases de veículos, que tenta otimizar a utilização do tempo disponível em cada ciclo, adaptando-se as necessidades reais de cada fase do cruzamento. Este procedimento de controle permite detetar o congestionamento do cruzamento em caso de se apresentar tal situação. Também, este trabalho contém o projeto de um controlador de sinais de trânsito baseado em um microcomputador, que emprega a lógica de controle proposta. O controlador pode se ajustar através do painel de controle, para dirigir cruzamentos que tenham desde duas até cinco fases de veículos, com a possibilidade de incluir como uma fase adicional, o serviço de passagem de pedestres. / [en] For the use in isolated veicular intersections, here is presented a logic for control of the time in veicular phases that tends to optimise the cicle time, based on the specific necessites of each phase. With this control`s producere it is possible to detect a veicular congestion at the intersection. In this paper it is also presented the desing of a microcomputer based traffic signal controller wich implements the control logic proposed. The controller can work at intersections that have from two to five veicular`s phases and has also the possibility to include a pedestrian phase, by setting those conditions in the control panel.