Les systèmes d'aide à la conduite gèrent un grand volume de données qui doivent être mises à jour régulièrement. Cependant, ces systèmes ne permettent, ni de les stocker, ni de les gérer d'une manière efficace. Pour ces raisons, nous proposons l'intégration d'un système de bases de données temps réel (TR) dans les systèmes d'aide à la conduite. Cela permet d'améliorer la tolérance aux fautes, de réduire le nombre de transactions et de réduire leur temps de réponse. La gestion d'un grand volume de données et leurs contraintes TR rend ces systèmes plus complexes, ce qui rend leur modélisation plus difficile. Pour remédier à cette complexité, nous avons proposé trois patrons de conception en nous basant sur un processus de création de patrons. Ce processus permet de définir les étapes à suivre pour déterminer les fonctionnalités et les exigences du domaine d'aide à la conduite, d'une part, et de définir les règles d'unification pour générer les diagrammes UML de classes et de séquence, d'autre part. Pour représenter ces patrons, nous avons proposé le profil UML-RTDB2, pour tenir compte : (i) de l'expression de la variabilité des patrons, (ii) de la représentation des contraintes TR et des aspects non fonctionnels et (iii) des éléments instanciés à partir des patrons lors de la modélisation d'une application cible. Une fois les patrons créés, ils peuvent être réutilisés par les concepteurs pour modéliser des systèmes spécifiques. Pour cela, nous avons proposé un processus de réutilisation pour guider les concepteurs d'applications lors de la réutilisation des solutions de patrons. Enfin, nous avons procédé à l'évaluation de ces patrons en utilisant deux catégories de métriques. / Advanced Driver Assistance Systems (ADAS) manage an important volume of data that must be updated regularly. However, ADAS don't store, nor manage efficiently these data. For these reasons, we propose to integrate a real-time (RT) database system into ADAS. The integration of the RT database system allows improving the fault tolerance, reducing the number of transactions and minimizing their response time. The management of a lot of data makes these systems complex, thus, their design is highly difficult. To tackle this problem, we have proposed three patterns based on the pattern development process. This process allows defining the steps to follow in order to determine the functionalities and the requirements of the driver assistance domain on one hand, and defining the unification rules for the generation of the UML class and sequence diagrams, on the other hand. In order to represent these patterns, we have proposed UML-RTDB2 profile, which allows (i) expressing the variability of patterns, (ii) representing the real time constraints and the non functional properties and (iii) identifying the role played by each pattern element in a pattern instance. Once the proposed patterns are created, they can be reused by designers to model a specific application. For this reason, we have proposed a process to assist the applications designers when instantiating the patterns solutions. Finally, we have evaluated these patterns based on two categories of metrics.

More than 900 bicyclists died in motor vehicle crashes in the United States in 2020, which represents a 50% increase from 2010 and the highest number of bicyclist deaths in nearly 35 years [1]. Reversing this trend will require efforts on multiple fronts, including reducing vehicle speeds and improving roadways and vehicles to be more hospitable to cyclists. Automatic emergency braking (ABB) with cyclist detection is a vehicle countermeasure with potential to prevent bicycle-motor vehicle crashes. AEB systems, which typically warn drivers of an impending collision and brake if drivers do not respond, have been shown to reduce vehicle-to-vehicle rear-end crash rates by 50% [2] and pedestrian crash rates by 27% [3]. Little is known about the real-world effects of ABB with cyclist detection on bicycle crashes. Subaru's EyeSight system, which includes ABB, has been capable of detecting cyclists in parallel configurations beginning in model year (MY) 2013 in the United States. The ability to detect cyclists in perpendicular configurations was added to some models beginning in MY 2022. The goal of this study is to evaluate the effects of the early version of EyeSight on U.S. bicycle crashes. [from Introduction]

Annually, road accidents cause more than 1.2 million deaths, 50 million injuries, and US$ 518 billion of economic cost globally. About 90% of the accidents occur due to human errors such as bad awareness, distraction, drowsiness, low training, fatigue etc. These human errors can be minimized by using advanced driver assistance system (ADAS) which actively monitors the driving environment and alerts a driver to the forthcoming danger, for example adaptive cruise control, blind spot detection, parking assistance, forward collision warning, lane departure warning, driver drowsiness detection, and traffic sign recognition etc. Unfortunately, these systems are provided only with modern luxury cars because they are very expensive due to numerous sensors employed. Therefore, camera-based ADAS are being seen as an alternative because a camera has much lower cost, higher availability, can be used for multiple applications and ability to integrate with other systems. Aiming at developing a camera-based ADAS, we have performed an ethnographic study of drivers in order to find what information about the surroundings could be helpful for drivers to avoid accidents. Our study shows that information on speed, distance, relative position, direction, and size & type of the nearby vehicles & other objects would be useful for drivers, and sufficient for implementing most of the ADAS functions. After considering available technologies such as radar, sonar, lidar, GPS, and video-based analysis, we conclude that video-based analysis is the fittest technology that provides all the essential support required for implementing ADAS functions at very low cost. Finally, we have proposed a Smart-Dashboard system that puts technologies – such as camera, digital image processor, and thin display – into a smart system to offer all advanced driver assistance functions. A basic prototype, demonstrating three functions only, is implemented in order to show that a full-fledged camera-based ADAS can be implemented using MATLAB. / Phone# 00966-56-00-56-471

Future heavy duty vehicles will be designed and manufactured with improved Advanced Driver Assistance Systems, ADAS. When developing ADAS, an accurate model of the vehicle dynamics greatly simplifies the development process. One element integral to the vehicle lateral dynamics and development of ADAS is the steering system. This thesis aims to develop an accurate model of a heavy duty vehicle steering system suitable for simulations. The input to the system is an input torque at the steering wheel and the output is the wheel angle. Physical models of the system components are developed using bond graphs and known relations. Some components are modelled with non-linear inefficiencies and friction of different complexity. Unknown parameters and functions are identified from measurement data using system identification tools such as, for example, linear regression and non-linear grid search. The different subsystems are identified separately to the extent deemed possible. Different model designs are considered, validated, and compared. The advantages and disadvantages of different model choices are discussed. Finally, a non-linear state space model is selected for its high accuracy and efficiency. As this final model can be used to simulate a heavy duty vehicle steering system on a desktop computer faster than real time, it fulfills its purpose.

Förarstödsystem har funnits länge, men tekniken och digitaliseringen fortsätter att utvecklas och så även förarstödsystemen. De system som utvecklas idag har större och mer fokuserad utveckling under kort tid och är för många okända system. Eftersom det finns begränsad kunskap om hur förarstödsystem fungerar när det kommer till detektering av vilt syftar denna studie till att skapa en fördjupad förståelse för vilka funktioner som finns i förarstödsystemen idag. Studien har skett genom en nulägesanalys baserad på hur systemen kan minska kollision med vilt samt vilka framtidsutsikter det finns beträffande utveckling av dessa system och funktioner. För att uppfylla syftet har intervjuer genomförts med ledande biltillverkarna och underleverantörer till fordonsindustrin och branschexperter. Insamlad empiri från intervjuer har kompletterats med ett urvalresultat i tidigare forskning. För att bearbeta insamlad empiri används en tematisk analys. Studiens bearbetade empiri presenteras i följande teman: Presentation av företag och respondenter; Nulägesanalys; Utmaningar och framtida möjligheter; Kommersiella aspekter; och Framtidsprojektioner. Det råder en samstämmig bild i resultatet kring att stödsystem för viltdetektion är i sin linda och att de främsta vinsterna med dessa är en tidig detektion av vilt, vilket kan minska risken för allvarliga olyckor. Tekniska framsteg görs vars syfte främst är att göra föraren mer medveten i sin körning och samtidigt utöka kommunikationen mellan trafikanter och infrastruktur. Hinder för utveckling ligger i marknadsmässiga faktorer och otydliga nationella och internationella regler. / Driver support systems have been around for a long time, but technology and digitization continue to evolve daily. Technology development is in constant motion. The systems that are being developed today have, in relation to the remaining vehicle technology, been given a significantly large and conditional development in a short time and are still new to humans. As there is limited knowledge about how driver support systems work when it comes to game detection, this study aims to create an in-depth understanding of what functions exist in driver support systems today, through a kind of current situation analysis, which detects game, how these can reduce collision with game and what prospects there are regarding the development of these systems and functions. To fulfil the purpose, interviews are conducted with leading car manufacturers and subcontractors to the automotive industry and industry experts. A thematic analysis is used to process collected empirical data. The study's processed empirics are presented in the following themes: Presentation of companies and respondents; Current situation analysis; Challenges and future opportunities; Commercial aspects; and Future projections. There is a consistent picture in the results that support systems for game detection are in their infancy and that the main benefits of these are an early detection of game, which can reduce the risk of serious accidents. Technical advances are being made whose main purpose is to make the driver more involved in their driving and at the same time expand the communication between road users and infrastructure. Obstacles to development lie in market factors and unclear national and international rules.

The aim of this PhD thesis was to investigate traffic safety and mobility for individuals with hearing loss (HL). Three studies were conducted: 1. a questionnaire survey aimed to evaluate differences in choice of transportation that might be related to HL, 2. a driving simulator study that looked into compensatory strategies and evaluated the efficiency of a tactile signal to alert the driver, and 3. a field study to evaluate these effects in real traffic and to evaluate a navigation system with a supportive tactile signal. The effects of HL discovered in this thesis add to the knowledge and understanding of the influence of HL on traffic safety and mobility. Differences found consistently point to a generally more cautious behavior. Compensatory and coping strategies associated with HL are bound to driving complexity and appear when complexity increases. These strategies include driving at lower speeds, using a more comprehensive visual search behavior and being less engaged in distracting activities. Evaluation of a tactile signal showed that by adding a tactile modality, some driver assistance systems can also be made accessible to drivers with HL. At the same time, the systems might be more effective for all users, since the driver can be more focused on the road. Based on the results in this thesis, drivers with HL cannot be considered an increased traffic safety risk, and there should be no need for adjustments of the requirements of hearing for a license to drive a car. / Syftet med den här doktorsavhandlingen var att undersöka trafiksäkerhet och mobilitet för individer med hörselnedsättning (HN). Tre studier har genomförts: 1. en enkätstudie för att undersöka skillnader i transportvanor relaterade till HN, 2. en körsimulatorstudie for att titta på kompensatoriska strategier och utvärdera effektiviteten i en taktil signal för att påkalla förarens uppmärksamhet och 3. en fältstudie för att undersöka effekterna i riktig trafik samt utvärdera ett navigationssystem med en taktil signal som stöd för navigering. Effekterna av HN som kom fram i denna avhandling bidrar till kunskapen och förståelsen för hur HN påverkar trafiksäkerhet och mobilitet. De funna skillnaderna pekar konsistent mot ett generelltmera försiktigt beteende. Kompensatoriska - och copingstrategier förknippade med HN beror på körkomplexitet och observeras när komplexiteten ökar. Dessa strategier innebär körning med lägre hastighet, mera heltäckande visuell avsökning och mindre engagemang i distraherande uppgifter. Utvärdering av en taktil signal visade att genom att lägga till en taktil modalitet kan vissa förarstödsystem bli tillgängliga även för förare med HN. Samtidigt kan systemen bli mera effektiva för alla användare eftersom föraren då kan fokusera mera på vägen. Baserat på resultaten i den här avhandlingen kan inte förare med HN betraktas som någon förhöjd risk och det bör därmed inte finnas något behov av att justera hörselkraven när det gäller körkortsinnehav.

Le travail de cette thèse a été consacré au développement d’une nouvelle méthode de détection pour un système anticollision par la mesure de trajectographie, ce qui pourrait contribuer aux systèmes d’aide à la conduite. Pour obtenir une haute probabilité de détection, nous avons choisi la solution de vidéo stéréoscopique coopérative : la coopération entre véhicules rend la détection plus facile et fiable. Il y a deux participants dans le système : les véhicules « porteurs du système » aussi bien que les « suiveurs », sont équipés de caméras stéréoscopiques, c’est à dire de deux capteurs d’image, appartenant à des familles technologique à haute cadence; les véhicules « cibles » sont équipés des feux à Leds modulés, dont la fréquence de modulation est déjà connue par les véhicules « suiveurs ». Après filtrage dans l’espace temporel, le système ne détecte que des signaux issus des feux modulés, ce qui réduit fortement l’information à traiter par rapport aux calculs de trajectographie traditionnels. La détection de feux modulés est donc réalisée par le filtrage par traitement numérique des images, qui est adapté à la fréquence de modulation recherchée. Pour cela, nous avons proposé 3 types de filtres adaptés à la fréquence de modulation et conçus de façon à rejeter au mieux les signaux de fond.Pour évaluer les performances tant en détection qu’en réjection des fausses alarmes, nous avons d’abord effectué des simulations numériques en prenant en compte des signaux artificiels, puis des calculs sur vrais signaux obtenus dans les expérimentations avec véhicule d’essai statique, puis roulant. Les roulages étaient de différentes vitesses, de 30km/h jusqu’à 100km/h, ce qui nous a permis d’analyser le signal issu du feu ainsi que le comportement de nos filtres à des vitesses angulaires de feu nulles, faibles ou élevées. Le résultat de ces expérimentations montre que le filtrage permet de détecter les feux à Leds de type DRL jusqu’à 140m sans aucune fausse détection sur le fond. Ces expérimentations sont une étape essentielle pour définir de façon plus précise un tel système, en particulier dans le choix du seuil. Nous avons aussi évalué des technologies qui peuvent améliorer la performance du système, mais qui ne sont pas encore prêtes à industrialiser. Par exemple, les « rétines » artificielles nous permettent d’utiliser les filtres analogiques intégrés, et ainsi de réduire leurs bandes passantes. / This thesis was devoted to the development of a new detection method for vehicular collision avoidance system based on trajectory measurement, which could contribute to driver assistance systems.In order to obtain high detection probability, we have chosen the cooperative stereoscopic video solution: the cooperation between vehicles makes it easier and more reliable when they aim to detect each other. There are two participants in the system: the “system carriers" vehicles, or the " followers" are equipped with stereoscopic cameras (two image sensors), who belong to high speed technology families; the "targets" vehicles are equipped with modulated LED lights, with the modulation frequency being already known by the "followers". After space-time filtering, the system detects the signals emitted bymodulated lights sources, which greatly reduces the amount of information to be processed comparing to traditional trajectory calculations methods. The detection of modulated light is achieved by filtering based on digital image processing, which is adapted to the desired modulation frequency. We have proposed three types of filters suitable for detecting the modulation at this frequency and at the same time for rejecting the background as well as possible.In order to be able to evaluate the performances of both detecting signals and rejecting false alarms, we first performed numerical simulations based on the model signals, then calculations on real signals acquired in static and driving experiments. The tested speeds were from 30km/h up to 100km/h, which allowed us to analyze the signals emitted from vehicle lights as well as the behavior of our filters under different angular velocities of the lights (zero, low and high). The result of these experiments showed that our method of filtering could detect LED-type DRL lights up to 140m without any false alarm. This is essential to define more precisely the values of thresholds of such systems. We have also evaluated technologies that are possible to improve system performance in the future, which are not yet ready to be used in industry productions. For example, artificial "retinas" could allow us to integrate analog filters in the chips, and thus to reduce bandwidth of the filters.

The objective of this dissertation is to develop a Multi-Camera Lane Departure Warning (MCLDW) system and a framework to evaluate it. A Lane Departure Warning (LDW) system is a safety feature that is included in a few luxury automobiles. Using a single camera, it performs the task of informing the driver if a lane change is imminent. The core component of an LDW system is a lane detector, whose objective is to find lane markers on the road. Therefore, we start this dissertation by explaining the requirements of an ideal lane detector, and then present several algorithmic implementations that meet these requirements. After selecting the best implementation, we present the MCLDW methodology. Using a multi-camera setup, MCLDW system combines the detected lane marker information from each camera's view to estimate the immediate distance between the vehicle and the lane marker, and signals a warning if this distance is under a certain threshold. Next, we introduce a procedure to create ground truth and a database of videos which serve as the framework for evaluation. Ground truth is created using an efficient procedure called Time-Slicing that allows the user to quickly annotate the true locations of the lane markers in each frame of the videos. Subsequently, we describe the details of a database of driving videos that has been put together to help establish a benchmark for evaluating existing lane detectors and LDW systems. Finally, we conclude the dissertation by citing the contributions of the research and discussing the avenues for future work.

Durch das I-FAS der TU Chemnitz wurde im Rahmen des AKTIV-Projektes eine Probandenstudie zur Akzeptanz von Systemausprägungen einer Aktiven Gefahrenbremsung (AGB) bei PKW durchgeführt. Unter Verwendung eines stehenden Hindernisses wurden sechs Systemausprägungen verglichen, die von den AGB-Partnern in zwei Versuchsträger implementiert wurden. Die sechs Systemausprägungen werden nahezu identisch bewertet, solange Probanden keine Vergleichsmöglichkeit zu anderen Systemausprägungen haben. Wenn es zu einem Fahrereingriff kommt, ist der Eingriffszeitpunkt des Fahrers unabhängig von der gefahrenen Systemausprägung.

This master‘s thesis describes the design and implementation of the system for detection and recognition of speed limit road signs. It focuses on the recognition of the red circular speed limit sign from the image data using the computer vision methods. Several methods were programmed and tested as a part of this thesis. In the final solution, the segmentation based on YCbCr color model is used. Detection of the circular sign and final classification is performed by template matching method. Algorithm for the tracking of the detected signs between frames of the video is used for better performance in real-time recognition. Application is developed using MATLAB and Simulink. The result is a simple driver assistance system prototype, which can be implemented in any computer with camera. The correct function of the algorithm was confirmed during a testing in a traffic.