La distraction du conducteur est un des facteurs importants à l’origine des accidents de la route. La détection de la distraction dans le contexte industriel et à faible coût conduit à privilégier des indicateurs reposant sur les capteurs déjà disponibles dans un véhicule série. Cependant, les systèmes actuels sont en général insuffisamment fiables, notamment parce que les grandeurs observées pour réaliser la détection sont assez éloignées du phénomène purement physiologique de distraction. L’approche étudiée ici a consisté à rajouter un modèle de comportement du conducteur (modèle cybernétique), rendant compte des fonctions perceptives et motrices support du contrôle latéral du véhicule. Les paramètres de ce modèle ont été estimés en procédant tour à tour à une identification par paquet de données d’entrée/sortie et à une identification récursive, cette dernière permettant de suivre continûment l'évolution paramétrique. Ensuite, trois approches ont été envisagées pour modéliser voire estimer l’état de distraction, considérant successivement la distraction comme une perturbation affectant les paramètres, la sortie ou l’entrée du modèle cybernétique du conducteur:Approche 1 - La distraction est modélisée comme une perturbation additive en sortie du modèle. Le couple produit par le conducteur est comparé au couple prédit par le modèle rendant compte de la conduite hors distraction. L’erreur de prédiction du couple constitue dans ce cadre le résidu dont la sensibilité à l’état de distraction du conducteur a été étudiée.Approche 2 - La distraction est modélisée par des perturbations multiplicatives, affectant certains paramètres du modèle. L’analyse des paramètres obtenus dans des phases de conduite avec et sans distraction a permis d’étudier leur capacité à rendre compte de la nature et de l’état de la distraction.Approche 3 - La distraction est modélisée comme une perturbation additive sur l’entrée du modèle. L’estimation de cette perturbation constitue un résidu également sensible à l’état de distraction. Les principes et algorithmes proposés pour estimer l’état de distraction ont été validés à partir de données expérimentales collectées pendant une campagne de tests effectuée sur un simulateur de conduite à base fixe, impliquant 35 conducteurs. Les conditions de test alternaient des phases de conduite normale et sujettes à des distractions de différentes natures : distractions cognitive, visuelle, visuomotrice et motrice. Les trois approches proposées donnent des résultats similaires et cohérents entre eux. / Distracted driving is one of the important factors that cause road accidents. The detection of the driver’s state of distraction in the industrial context and at low-cost leads to privilege the indicators based on sensors that are already available on the vehicle. However,current systems are generally not reliable enough, especially because the observed magnitudes to achieve detection are quite far from a purely physiological phenomenon distraction. This led us to propose solutions based on a cybernetic driver model that represent the visual and motor process involved in the lateral control of the vehicle. The parameters of this model have been estimated by conducting successively identification exploiting data packets and recursive identification, the latter allowing to track continuously the parametric evolution over time. Then, three approaches were considered to model or estimate the state of distraction, by modeling alternately thedistraction as a disturbance affecting parameters, the output or the input of the cybernetic model of the driver:Approach 1 - The distraction is modeled as an additive disturbance on the model output. The experimental output, the driver steering wheel torque, is then compared with the predicted steering wheel torque to generate the torque prediction error that is sensitive to the state of distraction.Approach 2 - The distraction is modeled as disturbances that affect the model parameters. The analysis of these parameters identified during normal and distracted driving periods showed that the parameters’ variation depends effectively on the driver’s state of distraction.Approach 3 - Distraction is modeled as an additive disturbance on the input of the model. The estimate of this disturbance is also a significant residue, sensitive to the state of distraction. The principles and algorithms proposed for estimating the state of distraction were validated using experimental data collected during a test campaign conducted on a fixed-base driving simulator, involving 35 drivers. The test conditions alternated normal driving phases and prone to distractions of various kinds: cognitive distractions, visual, visual-motor and motor. The three proposed approaches give similar and consistent results between them.

Cílem této práce je návrh a implementace řešení transparentního šifrování pro platformu Microsoft Windows. Řešení by mělo být propojitelné s produktem prevence proti úniku dat (DLP) a rozšiřovat jej. K implementaci byl využit framework Microsoft File System Minifilter Driver, s jehož pomocí je možné sledovat a upravovat přístup k jednotlivým souborům na externích zařízeních nebo discích za běhu systému. Soubory jsou zabezpečeny na pozadí tak, aby uživatel nebyl neovlivněn při práci. Ovladač zajišťuje, že uživatel vždy pracuje s rozšifrovanými daty. Dále bude také vyvinuta externí aplikace, která umožňuje uživateli přistoupit k zašifrovaným datům, aniž by musel být v síti, kde DLP produkt běží.

VIRTUAL PROTOTYPING has been generally adopted in product development in order to minimise the traditional reliance on testing of physical prototypes. It thus constitutes a major step towards solving the conflict of actual increasing development cost and time due to increasing customer demands on one side, and the need to decrease development cost and time due to increasing competition on the other. Particularly challenging for the off-road equipment industry is that its products, working machines, are complex in architecture. Tightly coupled, non-linear sub-systems of different technical domains make prediction and optimisation of the complete system’s dynamic behaviour difficult. Furthermore, in working machines the human operator is essential for the performance of the total system. Properties such as productivity, fuel efficiency, and operability are all not only dependent on inherent machine properties and working place conditions, but also on how the operator uses the machine. This is an aspect that is traditionally neglected in dynamic simulations, because the modelling needs to be extended beyond the technical system. The research presented in this thesis focuses on wheel loaders, which are representative for working machines. The technical system and the influence of the human operator is analysed, and so-called short loading cycles are described in depth. Two approaches to rule-based simulation models of a wheel loader operator are presented and used in simulations. Both operator models control the machine model by means of engine throttle, lift and tilt lever, steering wheel, and brake only – just as a human operator does. Also, only signals that a human operator can sense are used in the models. It is demonstrated that both operator models are able to adapt to basic variations in workplace setup and machine capability. Thus, a “human element” can be introduced into dynamic simulation of working machines, giving more relevant answers with respect to operator-influenced complete-machine properties such as productivity, fuel efficiency, and operability already in the concept phase of the product development process. / <p>ISRN/Report code: LiU-Tek-Lic 2005:44</p>

For this project, a Software-In-the-Loop (SIL) simulation tool is used at Scania (“VTAB” – Virtual Truck and Bus), which simulates the submodels of the mechanical vehicle components together with the real control units. The simulation tool contains the following submodels: Engine model, Drivetrain model, Drive cycle model, Restbus model, and Driver model. The simulated human driver submodel in the restbus model outputs two pedal control signals to the control unit, namely the gas and brake pedals. With these two pedal signals, the control unit decides the modes of mechanical vehicle components. This driver model needs to be reworked to obtain a better velocity following performance. Two controllers, fuzzy PI anti-windup and backward calculation, are implemented in the driver model and compared by the velocity tracking accuracy and the pedal switching frequency. In the comparison and analysis section, two different cycles and two weights of payload are simulated. The simulation results demonstrate that both controllers can improve the driver model’s velocity tracing accuracy. Further, the fuzzy PI anti-windup controller is better when considering pedal signals fluctuation frequency and implementation complexity. / För detta projekt används ett simuleringsverktyg Software-In-the-Loop (SIL) på Scania (“VTAB” - Virtual Truck and Bus), vilket simulerar submodellerna för de mekaniska fordonskomponenterna tillsammans med de verkliga styrenheterna. Simuleringsverktyget innehåller följande submodeller: Motormodell, Drivmotormodell, Drivcykelmodell, Restbusmodell och Drivermodell. Den simulerade submodellen för mänsklig förare i restbussmodellen kommer att sända två pedalsstyrsignaler till styrenheten, nämligen gas och broms. Med dessa två pedalsignaler kan styrenheten avgöra lägen av mekaniska fordonskomponenter. Denna drivrutinmodell måste omarbetas för att få en bättre hastighetsspårnings presentationsförmåga. Två styrenheter, fuzzy PI anti-windup och bakåtberäkning, implementeras i förarmodell och jämförs respektive med hastighetsspårningsnoggrannhet och pedalväxelfrekvens. I jämförelseoch analysavsnittet simuleras två olika cyklar och två nyttolast. Simuleringsresultaten visar att båda kontrollerna kan förbättra förarmodellens hastighetsspårningskapacitet. Vidare är fuzzy PI-anti-windup-kontroller bättre när man tar hänsyn till pedalsignalernas fluktueringsfrekvens och implementeringskomplexitet

Car following (CF) behaviour modelling has made significant progress in both traffic engi-neering and traffic psychology during recent decades. Autonomous vehicles (AVs) have been demonstrated to optimise traffic flow and increase traffic stability. Consequently, sever-al car-following models have been proposed based on various car following criteria, leading to a range of model parameter sets. In traffic engineering, Intelligent Driving Model (IDM) are commonly used as microscopic traffic flow models to simulate a single vehicle's behav-iour on a road. Observational data can be employed to parameter calibrate IDM models, which enhances their practicality for real-world applications. As a result, the calibration of model parameters is crucial in traffic simulation research and typically involves solving an optimization problem. Within the given context, the Nelder-Mead（NM）algorithm, particle swarm optimization (PSO) algorithm and genetic algorithm (GA) are utilized in this study for parameterizing the IDM model, using abundant trajectory data from five different road conditions. The study further examines the effects of various algorithms on the IDM model in different road sections, providing useful insights for traffic simulation and optimization.:Table of Contents CHAPTER 1 INTRODUCTION 1 1.1 BACKGROUND AND MOTIVATION 1 1.2 STRUCTURE OF THE WORK 3 CHAPTER 2 BACKGROUND AND RELATED WORK 4 2.1 CAR-FOLLOWING MODELS 4 2.1.1 General Motors model and Gazis-Herman-Rothery model 5 2.1.2 Optimal velocity model and extended models 6 2.1.3 Safety distance or collision avoidance models 7 2.1.4 Physiology-psychology models 8 2.1.5 Intelligent Driver model 10 2.2 CALIBRATION OF CAR-FOLLOWING MODEL 12 2.2.1 Statistical Methods 13 2.2.2 Optimization Algorithms 14 2.3 TRAJECTORY DATA 21 2.3.1 Requirements of Experimental Data 22 2.3.2 Data Collection Techniques 22 2.3.3 Collected Experimental Data 24 CHAPTER 3 EXPERIMENTS AND RESULTS 28 3.1 CALIBRATION PROCESS 28 3.1.1 Objective Function 29 3.1.2 Errors Analysis 30 3.2 SOFTWARE AND METHODOLOGY 30 3.3 NM RESULTS 30 3.4 PSO RESULTS 37 3.4.1 PSO Calibrator 37 3.4.2 PSO Results 44 3.5 GA RESULTS 51 3.6 OPTIMIZATION PERFORMANCE ANALYSIS 58 CHAPTER 4 CONCLUSION 60 REFERENCES 62

Dans un contexte où la consommation de carburant est un poste de coût de plus en plus important, la consommation et la vitesse moyenne d'un poids-lourds est l'une des caractéristiques clés estimées durant le développement des nouveaux véhicules. Ainsi, nous désirons créer différents modèles de conducteurs en termes de consommation et de vitesse moyenne, c'est à dire en termes de conduite rationnelle. Nous proposons une méthode en deux étapes: premièrement la reconnaissance des évènements de conduite grâce à des attributs de conduite. Puis la quantification de trois types de conduite différents sur ces évènements. Suite à ces deux étapes, nous pouvons implémenter nos résultats dans un outil qui crée différents modèles de conducteurs pour la simulation et le banc à rouleaux. Les écarts entre conducteurs se mesurent en termes de consommation de carburant et de vitesse moyenne. Le taux de classification des évènements s'étend de 74% à 91% selon le type d'évènements. Ces résultats sont dus à la nature même des données et aux similarités entre les classes, mais nous estimons que ces taux sont suffisants pour notre application. Nous obtenons également des corrélations prometteuses entre les attributs de conduite sélectionnés et l'indicateur de conduite rationnelle. Nous avons notamment porté notre étude sur les évènements classiques: les ronds-points, les péages et les arrêts. Les résultats de l'outil que nous avons développé sont pertinents. Nous pouvons désormais simuler différents types de chauffeurs. Sur nos essais en simulation, l'adaptation de seulement 10% des évènements d'un cycle découle sur un gain en consommation de 1.5% et une vitesse moyenne 3% plus élevée pour un conducteur efficace. Ces résultats sont encourageants, surtout que le travail à venir visera à augmenter la diversité des évènements couverts. / Realistic driver behavior modeling for energy efficiency estimation during heavy-trucks vehicles development Abstract: In the context where fuel consumption is a growing cost center, fuel consumption of a truck coupled with its average speed is one of the key vehicle characteristics that needs to be optimized and accurately estimated during the truck design process. Consequently, we aim to create different driver behavior models for testing trucks regarding fuel consumption and average speed issues, i.e., rational driving. We propose a two-step method to model more accurately driving behavior: first, the identification of driving events through driving features. Second, the quantification of three different driving behaviors on the recognized driving events. Then we implement our results in a tool that creates these different driving behaviors. The output of this tool is a cycle adapted to a driver type in terms of fuel consumption and average speed, and that can be used in simulation and on chassis-dynamometer. The classification of driving events reaches classification rates between 74% and 91% depending on the events. We believe that they are sufficient for our application due to the raw nature of driving events and the similarities between the different classes. We also obtain promising results concerning the correlation between driving features and rational driving index. We focus especially on typical events, namely roundabout on extra-urban roads, toll on highways and stop on urban roads. The results of the developed tool prove to be efficient since we can now simulate different driving behaviors. On our test run in simulation, adapting only 10% of the events of a cycle produces fuel savings of 1.5% and an average speed which is 3% faster for an efficient driver than a non-efficient driver. These results are promising and we need to implement other events in the future.

The accuracy in determined energy consumption and runningtime of trains, by means of computer simulation, is dependent upon the various models used. This thesis aims at developing validated models of running resistance, train and of a generaldriver, all based on full scale testing. A partly new simple methodology for determining running resistance, called by energy coasting method is developed and demonstrated. An error analysis for this methodis performed. Running resistance of high speed train SJ X2000, conventional loco hauled passenger trains and freight trains is systematically parameterised. Influence of speed, number of axles, axle load, track type, train length,and train configuration is studied. A model taking into account the ground boundary layer for determining the influence ofmeasured head and tail wind is developed. Different factors and parameters of a train, that are vital for the accuracy in computed energy consumption and runningtime are identified, analysed and finally synthesized into a train model. Empirical models of the braking and the traction system, including the energy efficiency, are developed for the electrical locomotive of typeSJ Rc4, without energy regeneration. Driver behaviour is studied for freight trains and a couple of driving describing parametersare proposed. An empirical model of freight train driver behaviour is developed from fullscale testing and observations. A computer program, a simulator, is developed in Matlabcode, making use of the determined runningresistance and the developed models of train and driver. The simulator calculates the energy consumption and running time ofa single train. Comparisons between simulations and corresponding measurements are made. Finally, the influence of driving on energy consumption and running time is studied and demonstrated in some examples. The main conclusions are that: The method developed for determining running resistanceis quite simple and accurate. It can be used on any train andon any track. The running resistance of tested trains includes some interesting knowledge which is partly believed to be new. Mechanical running resistance is less than proportional to the actual axle load. Air drag increases approximately linearly with train length and the effect of measured head and tail wind on the air drag can be calculated if the groundboundary layer is considered. The developed train model, including running resistance, traction, braking etc. is quite accurate, as verified for the investigated trains. The driver model together with the train model insimulations, is verified against measurements and shows good agreement for energy consumption and running time. It is recommended to use a driver model, when calculating energy consumption and running times for trains. Otherwise, the energy consumption will most likely be over-estimated.This has been demonstrated for Swedish ordinary freighttrains. / QC 20100526

This diploma thesis deals with the creation of a computation driver model. In the first part, there is an overview on driver models for longitudinal and lateral control. Next, driving maneuvres that could be selected for testing of driver model are described. In the practical part, there is created a computational driver model, whose task is to follow required path. The resulting model is tested on three driving maneuvers - steady turning, moose test and slalom. Finally, this model is tested on the passage of a real track. For all these tracks, a comparison is made and the success of the model is evaluated.

This thesis tackles the problem of predicting the collision risk for vehicles driving in complex traffic scenes for a few seconds into the future. The method is based on previous research using dynamic Bayesian networks to represent the state of the system. Common risk prediction methods are often categorized into three different groups depending on their abstraction level. The most complex of these are interaction-aware models which take driver interactions into account. These models often suffer from high computational complexity which is a key limitation in practical use. The model studied in this work takes interactions between drivers into account by considering driver intentions and the traffic rules in the scene. The state of the traffic scene used in the model contains the physical state of vehicles, the intentions of drivers and the expected behaviour of drivers according to the traffic rules. To allow for real-time risk assessment, an approximate inference of the state given the noisy sensor measurements is done using sequential importance resampling. Two different measures of risk are studied. The first is based on driver intentions not matching the expected maneuver, which in turn could lead to a dangerous situation. The second measure is based on a trajectory prediction step and uses the two measures time to collision (TTC) and time to critical collision probability (TTCCP). The implemented model can be applied in complex traffic scenarios with numerous participants. In this work, we focus on intersection and roundabout scenarios. The model is tested on simulated and real data from these scenarios. %Simulations of these scenarios is used to test the model. In these qualitative tests, the model was able to correctly identify collisions a few seconds before they occur and is also able to avoid false positives by detecting the vehicles that will give way. / Detta arbete behandlar problemet att förutsäga kollisionsrisken för fordon som kör i komplexa trafikscenarier för några sekunder i framtiden. Metoden är baserad på tidigare forskning där dynamiska Bayesianska nätverk används för att representera systemets tillstånd. Vanliga riskprognosmetoder kategoriseras ofta i tre olika grupper beroende på deras abstraktionsnivå. De mest komplexa av dessa är interaktionsmedvetna modeller som tar hänsyn till förarnas interaktioner. Dessa modeller lider ofta av hög beräkningskomplexitet, vilket är en svår begränsning när det kommer till praktisk användning. Modellen som studeras i detta arbete tar hänsyn till interaktioner mellan förare genom att beakta förarnas avsikter och trafikreglerna i scenen. Tillståndet i trafikscenen som används i modellen innehåller fordonets fysiska tillstånd, förarnas avsikter och förarnas förväntade beteende enligt trafikreglerna. För att möjliggöra riskbedömning i realtid görs en approximativ inferens av tillståndet givet den brusiga sensordatan med hjälp av sekventiell vägd simulering. Två olika mått på risk studeras. Det första är baserat på förarnas avsikter, närmare bestämt att ta reda på om de inte överensstämmer med den förväntade manövern, vilket då skulle kunna leda till en farlig situation. Det andra riskmåttet är baserat på ett prediktionssteg som använder sig av time to collision (TTC) och time to critical collision probability (TTCCP). Den implementerade modellen kan tillämpas i komplexa trafikscenarier med många fordon. I detta arbete fokuserar vi på scerarier i korsningar och rondeller. Modellen testas på simulerad och verklig data från dessa scenarier. I dessa kvalitativa tester kunde modellen korrekt identifiera kollisioner några få sekunder innan de inträffade. Den kunde också undvika falsklarm genom att lista ut vilka fordon som kommer att lämna företräde.

L’activité de conduite automobile prend place dans un environnement dynamique en constante évolution. Le conducteur doit progresser sur la route au moyen de son véhicule, tout en interagissant adéquatement avec l'environnement et les autres usagers. Pour réaliser cette tâche, le conducteur doit percevoir son environnement, interpréter les événements pour se représenter correctement la situation de conduite, anticiper ces changements, et prendre des décisions afin d'engager des actions sur le véhicule lui permettant d'atteindre les buts qu'il se fixe à court et long terme. A cet égard, la complexité et la diversité des processus perceptifs, cognitifs et sensori-moteurs requis pour la conduite automobile font de cette activité un objet d'étude particulièrement riche pour les sciences de la cognition.Pour étudier l'activité du conducteur automobile afin de la comprendre, l'expliquer et peut-être la prédire, les sciences cognitives se dirigent vers la modélisation de la cognition humaine. Cette démarche permet une représentation et une description plus ou moins fine du système cognitif du conducteur automobile. Cependant, un modèle de la cognition ne permet qu'une description théorique. Grâce à son implémentation informatique, il devient possible de simuler les théories utilisées et déployer numériquement celles mises en jeu dans la modélisation cognitive.Ce travail de thèse s'articule autour de la modélisation cognitive du conducteur automobile, de son implémentation informatique sur une plateforme de développement virtuel et de sa simulation au sein de cette plateforme. Le modèle théorique que nous avons implémenté est COSMODRIVE, en développement au laboratoire du LESCOT à l'IFSTTAR, et la plateforme de développement accueillant le modèle est SIVIC, développée au LIVIC. C'est dans ce contexte que nous nous sommes engagés dans le développement computationnel et informatique du modèle COSMODRIVE, afin de pouvoir simuler l'activité perceptive et cognitive du conducteur automobile. Pour cela, nous nous sommes limités à certains processus cognitifs primordiaux, comme les fonctions stratégiques (planification d'itinéraires et réalisation de plans stratégiques), ou les fonctions perceptives (exploration et intégration de l'information visuelle), les fonctions cognitives tactiques (construction de représentations mentales, intégration perceptivo-cognitive de l'information, structuration des connaissances de conduite, etc), ou encore les fonctions d'exécution d'actions (régulation courte par zones enveloppes ou par points de poursuite).Par l'implémentation informatique du modèle COSMODRIVE sur SIVIC, il devient possible "d'incarner numériquement" des théories cognitives et de les "opérationnaliser" pour formuler des hypothèses de recherche sous la forme de prédictions de performances que l'on pourra évaluer empiriquement auprès de conducteurs humains. Ces hypothèses formulées, nous avons conduit des expérimentations sur un simulateur de conduite que nous avons construit. Afin d'éprouver notre modèle théorique et informatique du conducteur, nous avons comparé les performances des conducteurs humains avec les prédictions issues de la simulation. Les résultats obtenus ont permis de valider cette approche et de confirmer l'intérêt de la simulation cognitive pour appréhender les activités mentales du conducteur automobile. / Driving activity takes place in a dynamic and constantly changing environment. The driver has to make his car evolving on the road while ensuring adequate interactions with its close environment and other road users. In order to perform this task, the driver has to perceive the environment he is evolving in, to interpret events in order to correctly understand the current driving situation, to be able to anticipate its evolution and take decisions regarding vehicle control in order to reach his short and long term goals safely. As a result, both complexity and variety of perceptual, cognitive and sensorimotor processes involved in the driving activity make it very rich context for cognitive sciences.The modeling of human cognition, a specific method which belongs to cognitive sciences field, has been chosen to study driver's activity aiming at understanding, explaining or even predicting it. This approach allows a representation and a description of the driver's cognitive system with different levels of granularity. Thus, such a model offers only a theoretical description. When implemented on a computer, it opens the way to the simulation allowing the digital deployment of the theories involved in the cognitive model design.This thesis is focused on cognitive modeling of car driver, its implementation and its simulation using a virtual platform. The theoretical model that we implemented is COSMODRIVE, developed at IFSTTAR - LESCOT laboratory and the implementation platform we used for this, named SIVIC, is developed at IFSTTAR - LIVIC.This is the context where we started the computational development of the COSMODRIVE model in order to simulate the perceptual and cognitive activity of car driver. Indeed, we chose to limit our implementation to some crucial cognitive processes such as strategic functions (route planning and strategic plans execution), perceptual functions (exploration and integration of visual information), cognitive tactical functions (construction of mental representations, perceptual and cognitive integration of information, structuring of driving knowledge, etc.), or executive functions of actions (short control loop by ''envelopes zones'' or pursuit points).Through computer simulation, we used the numerical model as an innovative tool for scientific investigation in the field of cognitive sciences: The numerical simulation of cognitive functions identified and modeled by COSMODRIVE allowed us to define experimental hypotheses which leed us to conduct experiments in a driving simulator that we have built. To test the theoretical model and computer of the car driver, we compared the performance of human drivers on one hand and the predictions issued from the simulation on the other hand. It opens innovative opportunities for the development and the use of cognitive modeling and simulation of car driver.