The goal of this research is to develop a method of objectively comparing and optimizing the performance of demand-responsive transportation systems in heterogeneous environments. Demand-responsive transportation refers to modes of transportation that do not follow fixed routes or schedules, including taxis, paratransit, deviated-route services, ride sharing as well as other modes. Heterogeneous environments are transportation environments in which streets do not follow regular patterns, passenger behavior is difficult to model, and transit schedules and layouts are non-uniform. An example of a typical heterogeneous environment is a modern suburb with non-linear streets, low pedestrian activity, and infrequent or sparse transit service. The motivation for this research is to determine if demand-responsive transportation can be used to improve customer satisfaction and reduce operating costs in suburban and low-density urban areas where fixed-route transportation may be inefficient. This research extends existing comparison and optimization techniques that are designed to work in homogeneous environments. Homogeneous environments refer to transportation systems where the streets follow regular and repeating patterns, passengers are evenly distributed throughout the system, and the transit system is easily modeled. The performance of systems with these characteristics can be approximated with closed-form analytical expressions representing passenger travel times, vehicle distances traveled, and other performance indicators. However, in the low-density urban areas studied in this research, the street patterns and transit schedules are irregular and passenger behavior is difficult to model. In these areas, analytical solutions cannot be found. Instead, this research develops a simulation-based approach to compare and optimize performance in these heterogeneous environments. Using widely-available route-planning tools, open-source transit schedules, and detailed passenger data, it is possible to simulate the behavior of transit vehicles and passengers to such an exacting degree that analytical solutions are not needed. A major technical contribution of this research is the development of a demand-responsive transportation simulator to analyze performance of demand-responsive systems in heterogeneous environments. The simulator combines several open-source tools for route planning with a custom-built demand-responsive vehicle and passenger-itinerary optimizer to simulate individual vehicles and passengers within a large system. With knowledge of the street network, the transit schedule, passenger locations, and trip request times, the simulator will output exact passenger transit times, passenger travel distances, vehicle travel distance, and other performance indicators for a particular transportation setup in a given area. The simulator is used to develop a method of comparing various demand-responsive and fixed-route systems. By predefining a set of performance indicators, such as passenger travel time and operating cost, the simulator can be used to ascertain the performance of a wide array of transportation systems. Comparing the weighted cost of each type of system permits a transportation engineer or planner to determine what type of system will provide the best results in a given area. The simulator is extended to assist in optimization of the demand-responsive transportation system layout. A key problem that needs to be solved when implementing a demand-responsive system is to determine the size, shape, and location of the demand-responsive coverage areas, i.e., the areas in which passengers are eligible for demand-responsive transportation. Using a particle swarm optimization algorithm and the simulation-based comparison technique, the optimal size and shape for a demand-responsive coverage area can be determined. The efficacy of the comparison and optimization techniques is demonstrated within the city of Atlanta, GA. It is shown that for certain areas of the city of Atlanta, demand-responsive transportation is more efficient than the currently implemented fixed-route system. Depending on the objective of the transportation planner, passenger satisfaction as well as operating costs can be improved by implementing a demand-responsive system in certain low-density areas. The techniques introduced in this research, and the simulation tool developed to implement those techniques, provide a repeatable, accurate, and objective method with which to optimize and compare demand-responsive transportation systems in heterogeneous environments.

As Intelligent Transportation Systems (ITS) become more prevalent, there is a need for a system capable of the rigorous evaluation of new ITS strategies for a wide variety of applications. Pre-deployment testing and fine-tuning of the system, performance evaluation, and alternatives analysis are all potential benefits that could be gained through the evaluation of ITS. Simulation, an increasingly popular tool for transportation analysis, would seem an ideal solution to this problem as it allows for the consideration of many scenarios that may be improbable or impossible to observe in the field. Also, simulation provides a framework that allows for the application of rigorous analysis techniques to the output data, providing an accurate and statistically significant conclusion. The difficulty is that many ITS strategies are difficult or impossible to implement in a simulated environment. The rapid nature of technology development and the complicated nature of many ITS solutions are difficult to emulate in simulation models. Furthermore, the emulation of a particular ITS solution is not guaranteed to provide the same result that the physical system would, were it subject to the same inputs. This study seeks to establish a framework for the analysis of advanced ITS applications through the use of Hardware-in-the-Loop Simulation (HILS), which provides a procedure for interfacing simulation models with real-world hardware to conduct analysis. This solution provides the benefits of both advanced ITS evaluation and simulation for powerful and accurate analysis. A framework is established that includes all the steps of the modeling process including construction, validation, calibration, and output analysis. This ensures that the process surrounding the HILS implementation is valid so that the results of the evaluation are accurate and defendable. Finally, a case study of the application of the developed framework to the evaluation, a real-world implementation of an advanced ITS application (SCATS in this case) is considered. The effectiveness of the framework in creating and evaluating a corridor using a simulation model wed to real-world hardware is shown. The results of the analysis show the power of this method when correctly applied and demonstrate where further analysis could expand upon the proposed procedure.

Thesis (M.S.)--University of Texas at El Paso, 2008. / Title from title screen. Vita. CD-ROM. Includes bibliographical references. Also available online.

Thesis (Ph. D.)--Civil and Environmental Engineering, Georgia Institute of Technology, 2005. / Leonard, John D., Committee Chair ; Goldsman, Dave, Committee Member ; Amekudzi, Adjo, Committee Member ; Hunter, Michael, Committee Member ; Dixon, Karen, Committee Member. Vita. Includes bibliographical references.

As important components of intelligent transportation system, lane detection and tracking (LDT) and lane departure warning (LDW) systems have attracted great interest from the computer vision community over the past few years. Conversely, lane markings recognition (LMR) systems received surprisingly little attention. This thesis proposed a real-time lane assisting framework for intelligent vehicles, which consists of a comprehensive module and simplified module. To the best of our knowledge, this is the first parallel architecture that considers not only lane detection and tracking, but also lane marking recognition and departure warning. A lightweight version of the Hough transform, PPHT is used for both modules to detect lines. After detection stage, for the comprehensive module, a novel refinement scheme consisting of angle threshold and segment linking (ATSL) and trapezoidal refinement method (TRM) takes shape and texture information into account, which significantly improves the LDT performance. Also based on TRM, colour and edge informations are used to recognize lane marking colors (white and yellow) and shapes (solid and dashed). In the simplified module, refined MSER blobs dramatically simplifies the preprocessing and refinement stage, and enables the simplified module performs well on lane detection and tracking. Several experiments are conducted in highway and urban roads in Ottawa. The detection rate of the LDT system in comprehensive module average 95.9% and exceed 89.3% in poor conditions, while the recognition rate depends on the quality of lane paint and achieves an average accuracy of 93.1%. The simplified module has an average detection rate of 92.7% and exceeds 84.9% in poor conditions. Except the conventional experimental methods, a novel point cluster evaluation and pdf analysis method have been proposed to evaluate the performance of LDT systems, in terms of the stability, accuracy and similarity to Gaussian distribution.

Intelligent transportation systems (ITS) have become a popular method to enhance the safety and efficiency of transportation. Pedestrians, as an essential participant of ITS, are very vulnerable in a traffic collision, compared with the passengers inside the vehicle. In order to protect the safety of all traffic participants and enhance transportation efficiency, the novel autonomous vehicles are required to detect pedestrians accurately and timely. In the area of pedestrian detection, deep learning-based pedestrian detection methods have gained significant development since the appearance of powerful GPUs. A large number of researchers are paying efforts to improve the accuracy of pedestrian detection by utilizing the Convolutional Neural Network (CNN)-based detectors. In this thesis, we propose a one-stage anchor-free pedestrian detector named Bi-Center Network (BCNet), which is aided by the semantic features of pedestrians' visible parts. The framework of our BCNet has two main modules: the feature extraction module produces the concatenated feature maps that extracted from different layers of ResNet, and the four parallel branches in the detection module produce the full body center keypoint heatmap, visible part center keypoint heatmap, heights, and offsets, respectively. The final bounding boxes are converted from the high response points on the fused center keypoint heatmap and corresponding predicted heights and offsets. The fused center keypoint heatmap contains the semantic feature fusion of the full body and the visible part of each pedestrian. Thus, we conduct ablation studies and discover the efficiency of feature fusion and how visibility features benefit the detector's performance by proposing two types of approaches: introducing two weighting hyper-parameters and applying three different attention mechanisms. Our BCNet gains 9.82% MR-2 (the lower the better) on the Reasonable setup of the CityPersons dataset, compared to baseline model which gains 12.14% MR-2 . The experimental results indicate that the performance of pedestrian detection could be significantly improved because the visibility semantic could prompt stronger responses on the heatmap. We compare our BCNet with state-of-the-art models on the CityPersons dataset and ETH dataset, which shows that our detector is effective and achieves a promising performance.

The emergence of Vehicular Ad-hoc Networks (VANETs) in the past decade has added a level of complexity to the modelling of Intelligent Transportation System (ITS) applications. In this paper, the Vehicular Network Integrated Simulator (VNetIntSim) is introduced as a new transportation network and VANET simulation tool by integrating transportation and VANET modelling. Specifically, it integrates the OPNET software, a communication network simulator, and the INTEGRATION software, a microscopic traffic simulation software. The INTEGRATION software simulates the movement of travellers and vehicles, while the OPNET software models the data exchange through the communication system. Information is exchanged between the two simulators as needed. The paper describes the implementation and the operation details of the VNetIntSim as well as the features it supports such as multiclass support and vehicle reuse. Subsequently, VNetIntSim is used to quantify the impact of mobility parameters (vehicular traffic stream speed and density) on the communication system performance considering Transmission Control Protocol (TCP) and User Datagram Protocol (UDP) applications. Specifically, the routing performance (packet drops and route discovery time), IP processing delay in case of a file transfer protocol (FTP) application, and jitter in case of a Voice over Internet Protocol (VoIP) application and evaluated.

Recently, according to people's requirements for safe and congestion-free driving in the public transportation system, the intelligent transportation system (ITS) has been widely concerned. To achieve a safe and time-saving driving experience in ITS, various data sharing methods are proposed to provide traffic information for drivers to perceive their surrounding driving environment. However, the high dynamic characteristic of the vehicular network (VNET) results in a challenging environment for establishing stable communication among vehicles. To face this challenge, a Cellular network-assisted Reliable Traffic-Aware Routing protocol (CRTAR) is proposed in this thesis to provide support for vehicle’s data routing process in a heterogeneous vehicular-cellular network environment. In the method, city-wide traffic information, i.e., traffic density and data transmission density of the road segments, is introduced into vehicle's data routing process to assist the vehicle in selecting the optimal data transmission route to deliver data packets. To further improve the stability of inter-vehicle communication, the link lifetime between vehicles is also considered to select the next forwarder that can establish relatively robust communication. CRTAR takes advantage of the reliability and low-latency features of the communication technology in the cellular network and combines the cellular network with VNET to achieve real-time and reliable Vehicle-to-Infrastructure (V2I) communication. Meanwhile, it realizes the Vehicle-to-Vehicle (V2V) communication by the Dedicated Short Range Communication (DSRC) to mitigate the overload of backbone resources caused by using the cellular network. To be specific, in the method, vehicles can request city-wide traffic information via the cellular network from a cloud service that is connected to the remote data center located in the traffic management agency without latency. According to the real-time traffic information, the source vehicle can execute the data routing process with a global view of the system to calculate the data transmission route that has sufficient transmission resources to the target vehicle. The source vehicle then transmits data to the target via the vehicles in the calculated transmission route. During the forwarding process, vehicles prefer to forward the data packet to the next vehicle with a longer link lifetime. Furthermore, effective backup and recovery strategies are designed for route maintenance. The effectiveness of CRTAR is further verified by conducting simulation experiments.

The purpose of this research is to explore the operational characteristics of a Maglev-based Automated Highway System and how it would interact with freeway operations. The extension of traditional traffic flow phenomenon, including weaving, merging, and stopping distance, into the automated system is looked at. These are also extended into platoon operations and their effect on such properties as gap control and ultimately the capacity of such a system. The ability to incorporate an AHS system into the existing Interstate Highway System is investigated. This includes placing the magways in the right-of-way of the highway system and interfacing the AHS with the existing freeways. A model is developed and run to simulate the assignment of traffic between the freeway and the guideway links. Both operational concepts of user equilibrium and system optimal conditions are explored, and equations are found to estimate the amount of traffic which can be found on the links based on the total traffic volume. / Master of Science

No description available.