Driver psychology is one of the most difficult phenomena to model in the realm of traffic flow theory because mathematics often cannot capture the human factors involved with driving a car. Over the past several decades, many models have attempted to model driver aggressiveness with varied results. The recently proposed Longitudinal Control Model (LCM) makes such an attempt, and this paper offers evidence of the LCM's usefulness in modeling road dynamics by analyzing deceleration rates that are commonly associated with various levels of aggression displayed by drivers. The paper is roughly divided into three sections, one outlining the LCM's ability to quantify forces between passive and aggressive drivers on a microscopic level, one describing the LCM's ability to measure aggressiveness of platoons of drivers, and the last explaining the meaning of the model’s derivative. The paper references some attempts to capture driver aggressiveness made by classic car-following models, and endeavors to offer some new ideas in study of driver characteristics and traffic flow theory.

Light field imaging, specifically synthetic aperture (SA) refocusing is a method used to combine images from an array of cameras to generate a single image with a narrow depth of field that can be positioned arbitrarily throughout the volume under investigation. Creating a stack of narrow depth of field images at varying locations generates a focal stack that can be used to find the location of objects in three dimensions. SA refocusing is particularly useful when reconstructing particle fields that are then used to determine the movement of the fluid they are entrained in, and it can also be used for shape reconstruction. This study applies SA refocusing to reacting flows and microscopic flows by performing shape reconstruction and 3D PIV on a flame, and 3D PIV on flow through a micro channel. The reacting flows in particular posed problems with the method. Reconstruction of the flame envelope was successful except for significant elongation in the optical axis caused by cameras viewing the flame from primarily one direction. 3D PIV on reacting flows suffered heavily from the index of refraction generated by the flame. The refocusing algorithm used assumed the particles were viewed through a constant refractive index (RI) and does not compensate for variations in the RI. This variation caused apparent motion in the particles that obscured their true locations making the 3D PIV prone to error. Microscopic PIV (µPIV) was performed on a channel containing a backward facing step. A microlens array was placed in the imaging section of the setup to capture a light field from the scene, which was then refocusing using SA refocusing. PIV on these volumes was compared to a CFD simulation on the same channel. Comparisons showed that error was most significant near the boundaries and the step of the channel. The axial velocity in particular had significant error near the step were the axial velocity was highest. Flow-wise velocity, though, appeared accurate with average flow-wise error approximately 20% throughout the channel volume.

Desorption electrospray ionization (DESI) is an ambient ionization technique for mass spectrometry. This solvent based desorption ion source has wide applicability in surface analysis with minimal sample preparation. Interest in improving detection limits, broadening applications, and increasing the spatial resolution for chemical imaging has led to studies of the DESI mechanism. An inverted microscope has been used to image interactions between the DESI spray and test analytes on a glass surface. Microscopic images recorded with millisecond time resolution have provided important insights into the processes governing analyte transport and desorption. These insights are the basis of a rivulet-based model for desorption that differs significantly from the widely-accepted momentum transfer model.

In nature, microscopic surface textures impact useful function, such as the drag reduction of shark skin (Dean & Bhushan, 2010) and superhydrophobicity of the lotus leaf(Pan, Kota, Mabry, & Tuteja, 2013). In this study, we explore these phenomena by re-creating microscopic surface textures via the method of interfacial flow instability in drying polyvinylidene fluoride (PVDF) acetone solutions. In general, PVDF films can be made using either spin coating or electrospray deposition with various weight concentrations in acetone. In order to study the morphology of the porous structure of PVDF films, wet deposition samples were fabricated by spin coating or near-field electrospray. Possible theories are discussed and examined to explain the formation of these porous structures resulting in development of a well-controlled method to create porous PVDF films with various pore sizes and pore densities. All samples are characterized and found to exhibit superhydrophobicity and drag reduction. To connect porous PVDF film morphology to the established field of dry particle fabrication, PVDF particle synthesis by far-field electrospray is also reviewed and discussed. An established method to generate polymer particles of different morphologies in other polymers (Almeria-Diez, 2012) by electrospray drying is confirmed using PVDF as well. Due to the ability of scalable and re-configurable electrospray, the microscopic surface textures can be applied to areas of any size to reduce drag or impart water-repelling properties.

A great deal of research has been conducted on Central Florida toll roads to better understand the characteristics of the tolling operation. In this thesis, the development and calibration of a toll plaza simulation models will be analyzed using two simulation programs varying mostly in their modeling theory. The two models utilized are, SHAKER, a deterministic queuing model for vehicles utilizing toll collection facilities, and VISSIM, a globally popular stochastic simulation software. The benefits of simulation models leads to the purpose of this thesis, which is to examine the effectiveness of two toll modeling programs that are similar in purpose but vary in approach and methodology. Both SHAKER and VISSIM toll plaza models have the potential to work as a tool that can estimate the maximum throughput and capacity of toll plazas. Major operational benefits resulting from developing these models are to simulate and evaluate how traffic conditions will change when demand increases, when and if queues increase when a lane is closed due to maintenance or construction, the impact of constructing additional lanes, or determining whether or not the best lane type configuration is currently implemented. To effectively calibrate any model available site data must be used to compare simulation results to for model validity. In an effort to correctly calibrate the SHAKER toll plaza tool and VISSIM model, an extensive field collection procedure was conducted at four Florida Turnpike operated toll facilities located in Central Florida. Each site differed from the others in terms of number of lanes, lane configuration, toll base fee, highway location, traffic demand, and vehicle percentage. The sites chosen for data collection were: the Lake Jesup Mainline Plaza along the Seminole Expressway (SR-417), the Beachline West Expressway Toll Plaza along the SR-528, the Daniel Webster Western Beltway Plaza along SR-429, and the Leesburg Toll Plaza along the Florida Turnpike Mainline SR-91. Upon completion of calibration of the two simulation models it is determined that each of the two software are successful in modeling toll plaza capacity and queuing. As expected, each simulation model does possess benefits over the other in terms of set up time, analysis reporting time, and practicality of results. The SHAKER model setup takes mere seconds in order to create a network and input vehicle, another few seconds to calibrate driving parameters, and roughly 10 additional seconds to report analysis. Conversely, setting up the VISSIM model, even for the most experienced user, can take several hours and the report analysis time can take several more hours as it is dependant on the number of required simulation runs and complexity of the network. VISSIM is most beneficial by the fact that its modeling allows for driver variability while SHAKER assumes equilibrium amongst lane choice and queuing. This creates a more realistic condition to observed traffic patterns. Even though differences are prevalent, it is important that in each simulation model the capacity is accurately simulated and each can be used to benefit operational situations related to toll plaza traffic conditions.

A key component of traffic models for simulating bicycle traffic focuses on capturing the interactions between cyclists and the cycling infrastructure. One of the most relevant features of the infrastructure that has a significant impact in bicycle traffic is the gradient of a bicycle path. Bicycle traffic simulations are a rather uninvestigated topic since historically, most focus on simulations has been on cars. However, bicycle simulations are an important tool to further investigate and understand cyclist’s behaviour. Therefore, the main objective of this thesis is to investigate and simulate free-riding behavior of cyclists in connection to the gradient, particularly on downhills. To do so, trajectory data of cyclists traveling on a downhill with a maximum gradient of 5.5\% are analysed to identify the impact of gradient on the speed and acceleration. The data received needed processing in order to be useful. This included filtering of the trajectories and excluding the data from cyclists which could not to be regarded as free-riding. As a result, a linear correlation is found between pedaling power and the gradient that can be used in microscopic bicycle traffic simulation. Based on this knowledge regarding this linearity, the approach used for modeling the gradient’s effect on the pedaling power is linear regression. The model can be developed in various ways, so instead of only choosing one model, several were developed and compared against each other. These models are then used for the simulation. The results indicate that the simulation captures well the impact of downhill gradients in a population of cyclists as it reproduces similar speed profiles to the ones observed. Therefore, it can be concluded that a power-based model is suitable for simulating free-riding behaviour of cyclists traveling in downhills. / En nyckelfaktor i trafikmodeller för cykelsimuleringar är att beskriva interaktionen mellan cykister och den omgivande infrastrukturen. En av de mest relevanta faktorerna hos infrastrukturen som har en signifikant påverkan på cykeltrafiken är lutningen på vägen. Simuleringar av cykeltrafik är ett tämligen outforskat område eftersom historiskt sett har mest fokus legat på simulering av bilar. Simuleringar är dock ett viktigt verktyg för att vidare undersöka cyklisters beteenden. Därmed är det huvudsakliga syftet med detta arbete att undersöka fria cyklisters beteenden med specifikt fokus på cyklister som cyklar i nedförsbackar. För att göra detta har data från cyklister som cyklar i nedförsbackar med en maximal lutning på 5.5\% analyserats för att analysera hur lutningen påverkar cyklisternas hastighet och acceleration. Den tillgängliga datan behövde bearbetning för att kunna användas. Detta innebar att filtrera datan samt att exkludera cyklister som ej kunde anses vara fria. Detta resulterade i att en linjär korrelation mellan effekten på pedalerna och lutningen på backen, denna korrelation kan användas för mikroskopisk cykelsimulering. Baserat på denna kunskap gällande linjäriteten kan linjär regressionsanalys användas för att modellera väglutningens inverkan på effekten från cyklisten. Modellen kan utvecklas på många olika sätt, och istället för att enbart välja en av dessa så utvecklades flera som kan jämföras med varandra. Dessa modeller används sedan för simuleringen. Resultaten indikerar att simuleringarna väl beskriver nedförsbackarnas påverkan på cyklister då simuleringarna genererar liknande hastighetskurvor som den uppmätta datan. Därmed kan slutsatsen dras att effektbaserade modeller är lämpliga för att simulera fria cyklisters beteenden när de cyklar i nedförsbackar.

Driving simulators are used to conduct experiments on for example driver behavior, road design, and vehicle characteristics. The results of the experiments often depend on the traffic conditions. One example is the evaluation of cellular phones and how they affect driving behavior. It is clear that the ability to use phones when driving depends on traffic intensity and composition, and that realistic experiments in driving simulators therefore has to include surrounding traffic. This thesis describes a model that generates and simulates surrounding vehicles for a driving simulator. The proposed model generates a traffic stream, corresponding to a given target flow and simulates realistic interactions between vehicles. The model is built on established techniques for time-driven microscopic simulation of traffic and uses an approach of only simulating the closest neighborhood of the driving simulator vehicle. In our model this closest neighborhood is divided into one inner region and two outer regions. Vehicles in the inner region are simulated according to advanced behavioral models while vehicles in the outer regions are updated according to a less time-consuming model. The presented work includes a new framework for generating and simulating vehicles within a moving area. It also includes the development of enhanced models for car-following and overtaking and a simple mesoscopic traffic model. The developed model has been integrated and tested within the VTI Driving simulator III. A driving simulator experiment has been performed in order to check if the participants observe the behavior of the simulated vehicles as realistic or not. The results were promising but they also indicated that enhancements could be made. The model has also been validated on the number of vehicles that catches up with the driving simulator vehicle and vice versa. The agreement is good for active and passive catch-ups on rural roads and for passive catch-ups on freeways, but less good for active catch-ups on freeways.

In September 2014, a new regulation was proposed in Sweden, which indicates aset of rules of giving way under different types of bicycle intersections. Meanwhile,new definitions were endued to bicycle passages and bicycle crossings. After thepriority of cyclists is guaranteed at bicycle crossings, potential delay for motorvehicles in the mainstream should not be ignored as well.Therefore, inspired by the previous study from Movea, this thesis project is goingto focus on unsupervised bicycle passages and bicycle crossings, with the aim ofassessing and inspecting the impacts of different bicycle intersections on vehicledelays.The relationship between delay and flows at bicycle intersections has beenresearched in detail. Initially, an empirical study has been carried out on the basisof data from field measurements. In addition, a microscopic traffic simulationmodel has been constructed for the analysis of theoretical situations. The delay isproved to have a linear relationship with vehicle flow and bicycle flow. However,the linear relationship based on field data turns out to be different from the onefrom simulation output.

Traffic shockwaves are a regularly occurring phe-nomenon in traffic that are a source of irritation and delaysfor the road users. One type of shockwave is the stop-and-gowave which forces entering drivers to stop and advance slowlyuntil the wave is passed. This project aims to design a controlalgorithm through the use of models and simulations to increasethe rate at which a stop-and-go wave dissipates. To design themodel and algorithm the Simulation of Urban MObility (SUMO)simulator and the Traffic Control Interface (TraCI) were usedin conjunction with Python. The setup used for simulation wasthat of a one way, two lane highway with an artificially inducedstop-and-go wave.The designed algorithm manages to dissipate a stop-and-go wavecompletely without introducing new ones. / Trafikvågorär ett vanligt förekommandefenomen i trafiken vilketär en orsak till frustration ochförseningar. En typ av vågär startochstop vågen som tvingarförare att stanna och långsamt fortsätta genom vågen tills denpasserat. Målet med detta projektär att utveckla en kontrol-lalgoritm med hjälp av modeller och simuleringar för attökaavtagandet av en sådan våg. För att utveckla modellen ochalgoritmen används simulatorn Simulation of Urban MObility(SUMO) och Traffic Control Interface (TraCI) i kombinationmed programmeringsspråket Python. Simulering gjordes på ettnätverk bestående av en enkelriktad, tvåfilig motorväg med enkonstgjord startochstop våg.En algoritm utvecklades som kan skingra en startochstop vågutan att skapa nya. / Kandidatexjobb i elektroteknik 2020, KTH, Stockholm

In this thesis, multiple on-road vehicle tracking problem is explored, with greater consideration of road constraints and interactions between vehicles. A comprehensive method for tracking multiple on-road vehicles is proposed by making use of domain knowledge of on-road vehicle motion. Starting with raw measurements provided by sensors, bias correction methods for sensors commonly used in vehicle tracking are briefly introduced and a fast but effective bias correction method for airborne video sensor is proposed. In the proposed method, by assuming errors in sensor parameter measurements are close to zero, the bias is separately addressed in converted measurements of target position by a linear term of errors in sensor parameter measurements. Based on this model, the bias is efficiently estimated by addressing it while tracking or using measurements of targets that are observed by multiple airborne video sensors simultaneously. The proposed method is compared with other airborne video bias correction methods through simulations. The numerical results demonstrate the effectiveness of the proposed method for correcting bias as well as its high computational efficiency. Then, a novel tracking algorithm that utilizes domain knowledge of on-road vehicle motion, i.e., road-map information and interactions among vehicles, by integrating a car-following model into a road coordinate system, is proposed for tracking multiple vehicles on single-lane roads. This algorithm is extended for tracking multiple vehicles on multi-lane roads: The road coordinate system is extended to two-dimension to express lanes on roads and a lane-changing model is integrated for modeling lane-changing behavior of vehicles. Since the longitudinal and lateral motions are mutually dependent, the longitudinal and lateral states of vehicles are estimated sequentially in a recursive manner. Two estimation strategies are proposed: a) The unscented Kalman filter combined with the multiple hypothesis tracking framework to estimate longitudinal and lateral states of vehicles, respectively. b) A unified particle filter framework with a specifically designed computationally-efficient joint sampling method to estimate longitudinal and lateral states of vehicles jointly. Both of two estimation methods can handle unknown parameters in motion models. A posterior Cramer-Rao lower bound is derived for quantifying achievable estimation accuracy in both single-lane and multi-lane cases, respectively. Numerical results show that the proposed algorithms achieve better track accuracy and consistency than conventional multi-vehicle tracking algorithms, which assumes that vehicles move independently of one another. / Thesis / Doctor of Philosophy (PhD)