Computer vision and image processing systems are ubiquitous in automotive domain and manufacturing industry. Lane detection warning systems has been an elementary part of the modern automotive industry. Due to the recent progress in the computer vision and image processing methods, economical and flexible use of computer vision is now pervasive and computing with images is not just for the realm of the science, but also for the arts and social science and even for hobbyists. Image processing is a key technology in automotive industry, even now there is hardly a single manufacturing process that is thinkable without imaging. The applications of image processing and computer vision methods in embedded systems platform, is an ongoing research area since many years. OpenCV, an open-source computer vision library containing optimized algorithms and methods for designing and implementing applications based on video and image processing techniques. These method are organized in the form of modules for specific field including, user-graphic interface, machine learning, feature extraction etc [43]. Vision-based automotive application systems become an important mechanism for lane detection and warning systems to alert a driver about the road in localization of the vehicle [1]. In automotive electronic market, for lane detection problem, vision-based approaches has been designed and developed using different electronic hardware and software components including wireless sensor, camera module, Field-Programmable Gate Array (FPGA) based systems, GPU and digital signal processors (DSP) [13]. The software module consists on the top of real-time operating systems and hardware description programming language including Verilog, or VHDL. One of the most time critical task of vision based systems is to test system applications in real physical environment with wide variety of driving scenarios and validating the whole systems as per the automotive industry standards. For validating and testing the advanced driver assistance systems, there are some commercial tools available including Assist ADTF from Elektrobit, EB company [43]. In addition to the design and strict real-time requirements for advanced driver assistance systems applications based on electronic components and embedded platform, the complexity and characteristics of the implemented algorithms are two parameters that need to be taken into consideration choosing hardware and software component [13]. The development of vision-based automotive application, based on alone electronic and micro-controller is not a feasible solution approach [35] [13] and GPU based solution are attractive but has many other issues including power consumption. In this thesis project, image and video processing module is used from OpenCV library for road lane detection problems. In proposed lane detection methods, low-level image processing algorithms and methods are used to extract relevant information for lane detection problem by applying contrast analysis at pixel level intensity values. Furthermore, the work at hand presents different approaches for solving relevant partial problems in the domain of lane detection. The aim of the work is to apply contrast analysis based on low-level image processing methods to extract relevant lane model information from the grid of intensity values of pixel elements available in image frame. The approaches presented in this project work are based on contrast analysis of binary mask image frame extracted after applying range threshold. A set of points, available in an image frame, based lane feature models are used for detecting lanes on color image frame captured from video. For the performance measurement and evaluation, the proposed methods are tested on different systems setup, including Linux, Microsoft Windows, CodeBlocks, Visual Studio 2012 and Linux based Rasbian-Jessie operating systems running on Intel i3, AMD A8 APU, and embedded systems based (Raspberry Pi 2 Model B) ARM v7 processor respectively.

A lane position system and enhancement techniques, for increasing the robustnessand availability of such a system, are investigated. The enhancements areperformed by using additional sensor sources like map data and GPS. The thesiscontains a description of the system, two models of the system and two implementedfilters for the system. The thesis also contains conclusions and results oftheoretical and experimental tests of the increased robustness and availability ofthe system. The system can be integrated with an existing system that investigatesdriver behavior, developed for fatigue. That system was developed in aproject named Drowsi, where among others Volvo Technology participated. / Ett filpositioneringssystem undersöks och förbättringstekniker för ökandet av robusthetoch tillgängligheten av ett sådant system genom att använda ytterligaresensorkällor som kartdata och GPS. Detta examensarbete presenterar beskrivningenav ett system, två modeller och två implementerade filter. Examensarbetetinnehåller också slutsatser och resultat av teoretiska och experimentella testersom plottar och grafer av ökad robusthet och tillgängligheten av systemet. Dettasystem kan bli integrerat med ett framtaget system som tittar på körrelaterat beteendevid trötthet. Systemet är utvecklat i ett projekt kallat Drowsi, där blandandra Volvo Technology deltog.

In recent years, autonomous driving has become an increasingly practical technology. With state-of-the-art computer and sensor engineering, autonomous vehicles may be produced and widely used for travel and logistics in the near future. They have great potential to reduce traffic accidents, improve transportation efficiency, and release people from driving tasks while commuting. Researchers have built autonomous vehicles that can drive on public roads and handle normal surrounding traffic and obstacles. However, in situations like lane changing and merging, the autonomous vehicle faces the challenge of performing smooth interaction with human-driven vehicles. To do this, autonomous vehicle intelligence still needs to be improved so that it can better understand and react to other human drivers on the road. In this thesis, we argue for the importance of implementing ”socially cooperative driving”, which is an integral part of everyday human driving, in autonomous vehicles. An intention-integrated Prediction- and Cost function-Based algorithm (iPCB) framework is proposed to enable an autonomous vehicles to perform cooperative social behaviors. We also propose a behavioral planning framework to enable the socially cooperative behaviors with the iPCB algorithm. The new architecture is implemented in an autonomous vehicle and can coordinate the existing Adaptive Cruise Control (ACC) and Lane Centering interface to perform socially cooperative behaviors. The algorithm has been tested in over 500 entrance ramp and lane change scenarios on public roads in multiple cities in the US and over 10; 000 in simulated case and statistical testing. Results show that the proposed algorithm and framework for autonomous vehicle improves the performance of autonomous lane change and entrance ramp handling. Compared with rule-based algorithms that were previously developed on an autonomous vehicle for these scenarios, over 95% of potentially unsafe situations are avoided.

The interest for autonomous driving assistance, and in the end, self-driving cars, has increased vastly over the last decade. Automotive safety continues to be a priority for manufacturers, politicians and people alike. Visual-based systems aiding the drivers have lately been boosted by advances in computer vision and machine learning. In this thesis, we evaluate the concept of an end-to-end machine learning solution for detecting and classifying road lane markings, and compare it to a more classical semantic segmentation solution. The analysis is based on the frame-by-frame scenario, and shows that our proposed end-to-end system has clear advantages when it comes detecting the existence of lanes and producing a consistent, lane-like output, especially in adverse conditions such as weak lane markings. Our proposed method allows the system to predict its own confidence, thereby allowing the system to suppress its own output when it is not deemed safe enough. The thesis finishes with proposed future work needed to achieve optimal performance and create a system ready for deployment in an active safety product.

O objetivo desta pesquisa foi recalibrar um modelo de simulação de rodovias de pista simples a fim de seja possível usá-lo no estudo de estradas do Estado de São Paulo. O modelo escolhido para recalibração é o TRARR (TRAffic on Rural Roads) versão 4.0, desenvolvido na Austrália pelo ARRB Transport Research Ltd. A recalibração do modelo consistiu em alterar parâmetros referentes ao desempenho (relação potência/massa) e características (massa, comprimento) dos veículos pesados e um parâmetro relativo ao comportamento dos veículos quando viajando em pelotões, de tal modo que os resultados da simulação ficassem compatíveis com os observados em campo. As medidas de desempenho escolhidas para a recalibração foram: a porcentagem de veículos trafegando em pelotões, a velocidade média dos veículos e a taxa de ultrapassagens. Os dados para a recalibração do modelo foram coletados num trecho de 2,8 km de extensão, na rodovia SP 255, entre as cidades paulistas de Araraquara e Ribeirão Preto. No processo de recalibração, as medidas de desempenho observadas no campo foram comparadas com as produzidas pelas versões recalibrada e original do simulador. Ao final da recalibração, pôde-se concluir que as duas versões do modelo são capazes de simular adequadamente o trecho estudado, sendo que o modelo recalibrado produz resultados ligeiramente melhores que os da versão original. O modelo recalibrado foi validado simulando-se um trecho de 10 km de extensão da SP 255 e comparando-se os valores medidos no trecho para as medidas de desempenho escolhidas com os valores produzidos pela versão recalibrada do simulador. Os resultados dessa comparação indicam que o modelo pode ser considerado capaz de simular adequadamente a operação de rodovias de pista simples no Estado de São Paulo. Os resultados do estudo também sugerem que a precisão pode ser melhorada com um ajuste fino de alguns dos parâmetros não modificados durante a recalibração. / The objective of this research is to recalibrate a two-lane highway simulation model to allow its employment within the context of the São Paulo state highways, Brazil. The model chosen for this recalibration is TRARR (TRAffic on Rural Roads) version 4.0, originally developed a in Australia by ARRB Transport Research Ltd. The procedure used for the model recalibration consisted in modifying input parameters related to heavy vehicle dimensions and performance characteristics (power/mass ratio, weight and length) as well as to the behavior of vehicles in platoons, in such a way that the model outputs are compatible with real traffic behavior. Three measures of performance were chosen for recalibration of the model: percent of jouney time spent following, average speed and overtaking rate. Data for model recalibration were collected at a 2.8 km rural section of SP 225, a highway which connects the cities of Ribeirão Preto and Araraquara in the state of São Paulo (Brazil). The recalibration procedure was based on comparisons of the measures of performance observed on the real highway with those produced by the simulator\'s recalibrated and original versions. At the end of the recalibration process, both the recalibrated and the original models were considered capable of reproducing the traffic behavior on a two-lane highway similar to the ones in the state of São Paulo. Furthermore, it was concluded that the recalibrated model is capable of producing results slightly better than the original one. The recalibrated model was submitted to a validation procedure in which a 10 km section of SP 255 was simulated and the measures of perfomance obtained with the model were compared to field observations. The results of the validation procedure indicate that recalibrated version of TRARR is capable of simulating traffic operations on rural two-lane highways in the state of São Paulo. The results of the research also suggest that the model results may be improved with fine-tuning of some of the parameters that were not modified during the recalibration procedure.

A necessidade de redução de custos logísticos tem obrigado as empresas a colaborar entre si. O problema de logística colaborativa aqui enfocado é assim definido: identificar ciclos (ou seja, um percurso fechado) em um conjunto de rotas de carga de lotação (onde o caminhão coleta carga em um ponto e vai diretamente ao local de descarga, pois é completamente preenchido) de vários embarcadores de forma a minimizar o reposicionamento (isto é, viagens sem carga útil) de caminhões, dado que o subconjunto de rotas de um determinado embarcador pode conter rotas que complementam aquelas de outro. Desta maneira, vários embarcadores combinados podem oferecer aos transportadores um conjunto de ciclos com movimentação regular de veículos com carga completa e com mínimo reposicionamento. Esse problema pode ser modelado como um problema particular de cobertura de conjuntos com restrição de ciclos, o problema de cobertura de rotas com cardinalidade restrita (PCRCR), que é NP-Hard. Este estudo apresenta uma heurística alternativa que obtém resultados, em média, 1,74% melhores que a literatura existente, além de solucionar instâncias maiores. Ademais, o tempo de execução da heurística cresce de forma polinomial em função do tamanho do problema, ao contrário dos demais métodos aqui avaliados, que possuem comportamento exponencial. / Cost and sustainability imperatives are compelling reasons to make companies to collaborate with each other in order to operate more efficiently. The shipper collaboration problem can be defined as how to identify tours (i.e. a closed path) in a set of lanes from various shippers that minimize truck repositioning (deadheads), as the sub-set of routes from a single shipper may have lanes that complement the routes of another shipper. Thus, combined shippers may offer to carriers a set of tours with regularly executed truckload movements (where the truck loads at a point and go directly to the disposal location) with minimum asset repositioning. This problem can be modeled as a particular case of the set covering formulation with constrained cycles, the cardinality constrained lane covering problem (CCLCP), which is NP-hard. This work resents an alternative heuristic that obtains results about 1.74% better than the existing literature, and solves larger instances. Besides, the heuristics execution time presents polynomial growth, unlike other methods that have exponential behavior.

O objetivo da tese foi adaptar, para as condições das rodovias brasileiras, as diretrizes da AASHTO para projetos de faixas adicionais de subida. Esta adaptação foi feita através de três aspectos: (i) geração de curvas de desempenho para caminhões típicos brasileiros; (ii) determinação da redução admissível de velocidade para caminhões; e (iii) estimativa do fluxo e da porcentagem de caminhões no tráfego que justifiquem economicamente a construção da faixa adicional. Quanto ao primeiro aspecto, constatou-se que as relações massa/potência encontradas para caminhões brasileiros (variam de 100 a 380 kg/kW) são significativamente maiores que a dos caminhões norte-americanos (120 kg/kW), o que inviabiliza o uso das curvas de desempenho que constam do livro verde da AASHTO. Nesta pesquisa, um modelo de locomoção de veículos foi calibrado e validado, logo após foram geradas curvas de desempenho para sete categorias de caminhões brasileiros. No segundo aspecto, as relações entre as diferenças de velocidade entre automóveis e caminhões, e índices de envolvimento de caminhões em acidentes foram obtidas. Para atingir esse objetivo, foram coletados dados sobre acidentes envolvendo caminhões, velocidade de veículos em aclives e volume diário médio. As relações foram comparadas com a utilizada nas diretrizes da AASHTO e as diferenças encontradas sugerem que a redução admissível na velocidade dos caminhões brasileiros deve ser de 20 km/h, em rodovias de pista simples, e de 35 km/h, em rodovias de pista dupla, se o índice de envolvimento de caminhões em acidentes for o que corresponde à redução máxima de velocidade admitida pela AASHTO (15 km/h). Em relação ao terceiro aspecto, foram analisadas relações benefício/custo para diversos cenários, visando estimar os fluxos de tráfego para os quais os custos fossem iguais aos benefícios. Os custos foram estimados a partir do investimento de capital necessário para a construção da ) faixa adicional, com base em projetos típicos do estado de São Paulo. Os benefícios estimados foram a redução do tempo de viagem e dos custos operacionais dos carros, ônibus e caminhões descarregados. Para a estimativa dos tempos de viagem, porcentagem do tempo trafegando em pelotões e porcentagem média de veículos em pelotões, foram feitas simulações, com e sem a faixa adicional, usando-se o modelo TRARR. Nessas simulações, os fluxos variaram entre 50 a 700 veíc/h e as porcentagens de caminhões de 10% a 40%. Os resultados indicam que os fluxos mínimos no aclive, que justifiquem economicamente a construção da faixa adicional, dependem da porcentagem de caminhões e da magnitude do aclive, variando de 340 veíc/h (10% de caminhões; aclives de até 3%) a 185 veíc/h (30% de caminhões; aclives de mais de 3%). / The objective of the research reported in this PhD dissertation was to adapt the AASHTO guidelines for design of climbing lanes to the Brazilian highways characteristics. This adaptation was carried out with regard to three aspects: (i) development of performance curves for typical Brazilian trucks; (ii) determination of the maximum speed reduction of slow trucks on upgrades; and (iii) estimation of minimum upgrade traffic flow rates and truck percents that would warrant the construction of climbing lanes on two-lane highways, from cost-effectiveness approach. Regarding the first aspect, it was verified that mass/power ratios for Brazilian trucks vary between 100 and 380 kg/kW and significantly differ from the ratio taken by AASHTO as representative of the average North-American truck (120 kg/kW); therefore, the truck performance curves from the AASHTO green book should not be used to design climbing lanes in Brazil. The green book\'s performance curves can be replaced by a family of performance curves for Brazilian trucks generated by a truck locomotion model, which was calibrated and validated in this research. With regard to the second aspect analyzed, the relationship between speed difference and truck accident involvement rates was studied. Data about truck accidents, speeds and annual average daily traffic were collected. These functions were compared to the one presented in the green book and the differences found suggest that, for the same truck accident involvement rate used to establish the maximum truck speed reduction adopted in the AASHTO policy (15 km/h), the admissible truck speed reduction on Brazilian roads would be 20 km/h on two-lane highways and 35 km/h on divided highways. With regard to the third aspect, benefit/cost ratios for several scenarios were analyzed in order to estimate upgrade traffic flow rates that would warrant the provision of climbing lanes. (continue) (continuation) The costs were estimated on the basis of the capital costs for the construction of the climbing lane, based on costs of climbing lanes built in the state of São Paulo. The benefits were estimated from the reduction of travel time and vehicle operating costs provided by the construction of a climbing lane. Travel times, percent time spent following, and average fraction of vehicles travelling in platoons with and without the climbing lane were estimated by means of traffic flow simulations carried out with TRARR. The simulations were run using traffic flows ranging from 50 to 700 veh/h, with truck percents varying from 10% to 40%. The results indicated upgrade traffic flow rates that warrants climbing lanes varying with truck percents and grades magnitude and range from 340 veh/h (grades of up to 3%, 10% trucks in flow) to 185 veh/h (grades of more than 3%, with 30% trucks).

Empresas estão procurando reduzir seus custos e aumentar seu desempenho e competitividade. Neste cenário de redução de custos, a logística colaborativa pode ser uma aliada. Numa rede complexa, onde embarcadores muitas vezes nem sabem da existência de outros embarcadores com demandas complementares, existe um potencial de sinergia e redução de custos através da diminuição de deslocamentos de veículos sem carga, ou seja, deslocamentos para reposicionar os veículos. Visando essa redução, o Problema de Cobertura de Rotas (PCR), que tem como objetivo cobrir rotas no mínimo custo, une as demandas de frete de vários embarcadores e tenta minimizar os deslocamentos sem cargas (reposicionamentos), reduzindo assim o custo total de toda a rede dos embarcadores envolvidos. Esta pesquisa propõe um modelo e uma heurística para resolver, em grande escala através de programação paralela, uma expansão do PCR. / Companies are looking to reduce costs and improve performance and competitiveness. In this cost-cutting scenario, collaborative logistics can be an ally. In a complex network where shippers often do not know of the existence of other shippers with complementary demands, there is potential for synergy and cost savings by reducing unloaded travelling of vehicles, i.e, the distance and time to reposition the vehicles\'. Towards that reduction, the Lane Covering Problem (LCP), which aims to cover at least cost routeslinks the various shippers\' demands of freight and tries to minimize operations without loads (repositioning), thus reducing the total cost of the entire network for involved shippers. This research proposes a model and an heuristic to solve, in large-scale through parallel programming, an expansion of the PCR.

Digitized by Kansas Correctional Industries

Esta dissertação de mestrado apresenta uma proposta de critério para a construção de faixas adicionais em rodovias com longos declives. A proposta contempla três aspectos: 1-) Elaboração de um programa de computador que permita calcular a temperatura do sistema de freios de qualquer veículo que esteja operando em qualquer ponto de uma rampa descendente, conhecidas as características do veículo, da massa transportada, da velocidade desenvolvida pelo veículo e da rampa em questão (comprimento e declividade). Em outras palavras: o programa permite gerar as curvas de desempenho para qualquer veículo de carga, em qualquer trecho de descida de uma rodovia; 2-) Determinação da redução de velocidade para caminhões nos trechos estudados; e 3-) Cálculo da porcentagem de veículos de carga no fluxo total de veículos que utilizam a rodovia, valor esse que viabilize economicamente a construção de faixas auxiliares. Esta dissertação aborda mais profundamente o primeiro aspecto da proposta. Os outros dois, devido à sua complexidade, são sugeridos como tema para trabalhos futuros. De posse do programa computacional, o mesmo foi utilizado para simular o comportamento da frota de veículos de carga brasileiros, em greides com comprimento variável de 1.000 a 12.000 metros e inclinações que variam de 1 a 12%. Também foi verificado o impacto da aplicação da resolução 777/93 do CONTRAN - que regulamenta o desempenho e o sistema de freios dos veículos de carga - na frota brasileira de veículos de carga. Os resultados obtidos indicam, além da importância de uma legislação adequada para o tema, uma urgente renovação ou adequação da frota brasileira de veículos de carga, haja visto que grande parte da mesma, por possuir uma elevada idade média, não atinge os requisitos mínimos para circular pelas rodovias brasileiras sem comprometimento das condições de segurança, dirigibilidade, estabilidade e desempenho. / This master thesis presents a criteria\'s proposal to construction of auxiliary lanes in roads with long and steep downgrades. The proposal contemplates three aspects: 1-) Elaboration of a computer\'s program that allows to obtain the brake system\'s temperature of any heavy truck performing at any point of a long and steep downgrade. It\'s necessary to know the truck\'s brake characteristics, the brake-motor power, the total mass transported, the vehicles\'speed and the downhill characteristics (length and percent). With this data, the program calculates the performance curves of brazilian trucks, regarding brakes\'system\'s termical aspects; 2-) Determination of the maximum speed reduction of slow trucks on downhill; and, 3-) Estimation of minimum downgrade traffic flow rates and truck percents that would warrant the construction of declivity help lanes, from costs-effectiveness approach. This master thesis broaches more profundity the first proposal\'s aspects. The aspects two and three, due to its complexity, are suggested to be future studies\'object. The computational program developed in this work was used to simulate the performance of brazilian\'s fleet heavy truck in declivities with lengths from 1.000 till 12.000 meters and declivity that vary from 1 till 12%. It was used, too, to show the impact of the resolution number 777/93 - edited by CONTRAN - applicated in the same fleet. This resolution rules ship vehicles\'brakes\'system and their performance. The obtain results denote, farther importance of a adequate legislation for the theme, a urgent renovation or adequation of brazilian\'s fleet heavy truck due to, a large part of it, with more than ten year of age, is unable to reach the minimum requisites to moving on brazilian roads and highways without implication of security, conductibility, stability and performance\'s conditions.