Le trafic routier est devenu de plus en plus intense. Une telle situation avec le manque de prudence des piétons constituent deux causes majeures de l’augmentation des accidents routiers. En France, 16% des accidents de la route en 2016 impliquent au moins un piéton et chaque année, environ de 800 piétons sont tués dans un accident de la circulation. De plus, la part des plus de 65 ans dans la mortalité piétonne est en hausse de 13% entre 2014 et 2016. Ainsi, par ce projet de thèse nous proposons une approche probabiliste pour inférer le type de comportement (à risque ou sécurisé) des piétons lors de la traversé de la rue. Cette approche se compose de 2 couches principales : Une couche basse, basée sur les techniques de vision par ordinateur, pour la collecte des paramètres des piétons, du trafic et des aménagements urbains et une couche haute, basée sur le Réseau Bayésien (RB), pour l’inférence du type de comportement. Plusieurs contributions et améliorations sont proposées pour la construction d’une telle approche que ce soit au niveau de la couche basse (techniques de détection et de suivi utilisées) ou au niveau de la couche haute (gestion des incertitudes des capteurs de vision et la mise en relation des paramètres hétérogènes et variées). / Road traffic has become more and more intense. Such as situation with thelack of attention of pedestrians are the two major causes of the increase in road accidents. In France, 16% of road accidents in 2016 involve at least one pedestrian and each year about 800 pedestrians are killed in a traffic accident. In addition, the proportion of pedestrian deaths having over 65 years old is up to 13% between 2014 and 2016. In fact, this thesis proposes a probabilistic approach to recognize pedestrians’ behavior (risky or secure) when crossing the street. This approach consists of two main layers: a low layer, based on computer vision techniques, for collecting pedestrian and traffic parameters and a high layer, based on the Bayesian Network, for behavior prediction. Several contributions and improvements are proposed for the construction of such an approach, whether at the level of the low layer (detection and tracking process) or at the level of the upper layer (management of the uncertainties of the vision sensors and the correlation of heterogeneous and varied parameters).

New cooperative intelligent transportation system (C-ITS) applications become enabled thanks to advances in communication technologies between vehicles(V2V) and with the infrastructure (V2I). Communicating vehicles share information with each other and cooperate, which results in improved safety, fuel economy, and traffic efficiency. An example of a C-ITS application is platooning, which comprises a string of vehicles that travel together with short inter-vehicle distances (IVDs). Any solution related to C-ITS must comply with high safety requirements in order to pass standardization and be commercially deployed. Furthermore, trusted safety levels should be assured even for critical scenarios. This thesis studies the conditions that guarantee safety in emergency braking scenarios for heterogeneous platooning, or string-like, formations of vehicles. In such scenarios, the vehicle at the head of the string emergency brakes and all following vehicles have to automatically react in time to avoid rear-end collisions. The reaction time can be significantly decreased with vehicle-to-vehicle (V2V) communication usage since the leader can explicitly inform other platooning members about the critical braking. The safety analysis conducted in the thesis yields computationally efficient methods and algorithms for calculating minimum inter-vehicle distances that allow avoiding rear-end collisions with a predefined high guarantee. These IVDs are theoretically obtained for an open-loop and a closed-loop configurations. The former implies that follower drives with a constant velocity until braking starts, whereas in the latter, an adaptive cruise control (ACC) with a constant-distance policy serves as a controller. In addition, further optimization of inter-vehicle distances in the platoon is carried out under an assumption of centralized control. Such an approach allows achieving better fuel consumption and road utilization. The performed analytical comparison suggests that our proposed V2V communication based solution is superior to classical automated systems, such as automatic emergency braking system (AEBS), which utilizes only onboard sensors and no communication. Wireless communication, enabling to know the intentions of other vehicles almost immediately, allows for smaller IVDs whilst guaranteeing the same level of safety. Overall, the presented thesis highlights the importance of C-ITS and, specifically, V2V in the prevention of rear-end collisions in emergency scenarios. Future work directions include an extension of the obtained results by considering more advanced models of vehicles, environment, and communication settings; and applying the proposed algorithms of safety guaranteeing to other controllers, such as ACC with a constant time headway policy.

Kraven på säkerhet och effektivitet ökar ständigt inom fordonsindustrin. För att uppfylla dessa strävar fordonstillverkare efter att uppnå en högre grad av autonomi, detta innebär dock att många problem måste lösas. Denna rapport behandlar ett av dessa, autonoma fordons möjlighet att identifiera utryckningsfordon. Målet är att presentera ett förslag på vilken teknik som anses mest lämpad för autonoma fordon att kommunicera med utryckningsfordon. Arbetet grundade sig i en förstudie där standarden ITS G5, IEEE 802.11g, ZigBee samt mobilnät analyserades utifrån deras tekniska specifikationer. Utifrån analysen presenterades tre situationer där de olika teknikernas användning ansågs begränsade, i tunnlar, i tät trafik samt på långa avstånd vid höga hastigheter. Dessa situationer ställde krav på teknikerna inom bland annat svarstid, räckvidd, överföringsförmåga samt möjlighet till direktkommunikation mellan fordonen. Utifrån dessa krav ställdes en jämförelsematris upp där de olika teknikernas prestanda jämfördes. Resultatet visar att ITS G5 och ZigBee har bäst prestanda på egen hand medan en kombination av mobilnät och ITS G5 uppnår högst prestanda. / The demand for safety and effectivity continuously increases within the automotive industry. One way to meet these demands is to achieve a higher level of autonomy, but to achieve the highest levels of autonomy there is a few problems to be solved along the way. This report treats one of these, an autonomous vehicle’s ability to identify emergency vehicles. The report, based on a pilot study, analyses the ITS G5 standard, IEEE 802.11g, Zigbee and mobile networks based on their technological specifications. From the analysis three situations are identified where the technologies are considered limited. These limitations are, but not limited to, reach, latency, data rates and ability to communicate vehicle to vehicle (V2V). The four technologies are then compared by these limitations in a matrix. The result shows that ITS G5 and ZigBee has the best performance by its own but the combination of mobile networks and ITS G5 shows the highest possible performance.

Evaluating the challenges and opportunities of cooperative autonomous vehicles (CAV) require an adapted simulation methodology reproducing realistic driving and sensory contexts. In this paper, we propose a RounD-like CARLA scenario reproducing in CARLA the driving context recorded in the RounD dataset. We focus in particular on roundabout scenarios, as they are considered particularly challenging for CAV. We present the methodology followed to generate the CARLA scenario and describe challenges to reproduce trajectories corresponding to RounD. Origin and destination of vehicles, waypoint and speed are extracted from RounD for CARLA vehicles to closely reproduce the driving patterns observed in RounD. The benefit of such scenario are manyfold, such as evaluating control algorithms of CAVs, deep AI reinforcement learning, or vehicular sensor data sampling under realistic driving contexts. It notably will reduce the gap of AI mechanisms for CAV between simulation scenarios and realistic conditions.

Vetiver grass or Chrysopogon zizanioides (L.) Roberty (1960) is a sterile grass which can regenerate vegetatively from clumps of the rootstock. This, as well as its vigorous and deep root system and flood tolerance makes it an ideal candidate for the use in soil remediation and erosion control. In South Africa, Hydromulch (Pty) Ltd. is part of the landscape, soil reclamation and erosion control industry. The company uses vetiver grass on a wide scale and has accumulated a collection of isolates to serve as possible germ lines for industrial use. Due to the different approaches in environmental management as well as environmental factors, a variety of ecotypes form during the planting and acclimatisation of this genus. Chrysopogon nigritanus (Benth.) Veldkamp (1999), which is a native species to Africa, is closely related to C. zizanioides and differs only slightly from C. zizanioides on a morphological level. The major difference between the two species is that C. nigritanus is able to seed freely and thus the use of this species should be avoided. The need arose to screen other non-fertile plants to uncover additional genotypic variety to enable diversification of vetiver plantings. The aim of this study was to characterise the genotype of 19 isolates of vetiver obtained from Hydromulch (Pty) Ltd. via sequencing analyses of three DNA fragments, ITS, ndhF and rbcL. In addition, the radial root anatomy was also investigated and compared with the genetic analyses. According to the results generated during this study, very little or no genotypical differences exist amongst the different isolates available from the Hydromulch (Pty) Ltd. plant collection. Only in the case of the ITS inference were differences observed between three of the studied isolates. There was no significant difference between the different isolates based on the root anatomy, with the exception of two of the studied isolates which formed starch granules. / MSc (Botany), North-West University, Potchefstroom Campus, 2014

Vetiver grass or Chrysopogon zizanioides (L.) Roberty (1960) is a sterile grass which can regenerate vegetatively from clumps of the rootstock. This, as well as its vigorous and deep root system and flood tolerance makes it an ideal candidate for the use in soil remediation and erosion control. In South Africa, Hydromulch (Pty) Ltd. is part of the landscape, soil reclamation and erosion control industry. The company uses vetiver grass on a wide scale and has accumulated a collection of isolates to serve as possible germ lines for industrial use. Due to the different approaches in environmental management as well as environmental factors, a variety of ecotypes form during the planting and acclimatisation of this genus. Chrysopogon nigritanus (Benth.) Veldkamp (1999), which is a native species to Africa, is closely related to C. zizanioides and differs only slightly from C. zizanioides on a morphological level. The major difference between the two species is that C. nigritanus is able to seed freely and thus the use of this species should be avoided. The need arose to screen other non-fertile plants to uncover additional genotypic variety to enable diversification of vetiver plantings. The aim of this study was to characterise the genotype of 19 isolates of vetiver obtained from Hydromulch (Pty) Ltd. via sequencing analyses of three DNA fragments, ITS, ndhF and rbcL. In addition, the radial root anatomy was also investigated and compared with the genetic analyses. According to the results generated during this study, very little or no genotypical differences exist amongst the different isolates available from the Hydromulch (Pty) Ltd. plant collection. Only in the case of the ITS inference were differences observed between three of the studied isolates. There was no significant difference between the different isolates based on the root anatomy, with the exception of two of the studied isolates which formed starch granules. / MSc (Botany), North-West University, Potchefstroom Campus, 2014

This research focused on a formal (theory based) approach to designing Intelligent Tutoring System (ITS) authoring tool involving two specific conventional pedagogical theories—Conversation Theory (CT) and Cognitive Apprenticeship (CA). The research conceptualised an Augmented Conversation and Cognitive Apprenticeship Metamodel (ACCAM) based on apriori theoretical knowledge and assumptions of its underlying theories. ACCAM was implemented in an Intelligent Learning Activity Builder System (ILABS)—an ITS authoring tool. ACCAM’s implementation aims to facilitate formally designed tutoring systems, hence, ILABS―the practical implementation of ACCAM― constructs metamodels for Intelligent Learning Activity Tools (ILATs) in a numerical problem-solving context (focusing on the construction of procedural knowledge in applied numerical disciplines). Also, an Intelligent Learning Activity Management System (ILAMS), although not the focus of this research, was developed as a launchpad for ILATs constructed and to administer learning activities. Hence, ACCAM and ILABS constitute the conceptual and practical contributions that respectively flow from this research. ACCAM’s implementation was tested through the evaluation of ILABS and ILATs within an applied numerical domain―the accounting domain. The evaluation focused on the key constructs of ACCAM―cognitive visibility and conversation, implemented through a tutoring strategy employing Process Monitoring (PM). PM augments conversation within a cognitive apprenticeship framework; it aims to improve the visibility of the cognitive process of a learner and infers intelligence in tutoring systems. PM was implemented via an interface that attempts to bring learner’s thought process to the surface. This approach contrasted with previous studies that adopted standard Artificial Intelligence (AI) based inference techniques. The interface-based PM extends the existing CT and CA work. The strategy (i.e. interface-based PM) makes available a new tutoring approach that aimed fine-grain (or step-wise) feedbacks, unlike the goal-oriented feedbacks of model-tracing. The impact of PM—as a preventive strategy (or intervention) and to aid diagnosis of learners’ cognitive process—was investigated in relation to other constructs from the literature (such as detection of misconception, feedback generation and perceived learning effectiveness). Thus, the conceptualisation and implementation of PM via an interface also contributes to knowledge and practice. The evaluation of the ACCAM-based design approach and investigation of the above mentioned constructs were undertaken through users’ reaction/perception to ILABS and ILAT. This involved, principally, quantitative approach. However, a qualitative approach was also utilised to gain deeper insight. Findings from the evaluation supports the formal (theory based) design approach—the design of ILABS through interaction with ACCAM. Empirical data revealed the presence of conversation and cognitive visibility constructs in ILATs, which were determined through its behaviour during the learning process. This research identified some other theoretical elements (e.g. motivation, reflection, remediation, evaluation, etc.) that possibly play out in a learning process. This clarifies key conceptual variables that should be considered when constructing tutoring systems for applied numerical disciplines (e.g. accounting, engineering). Also, the research revealed that PM enhances the detection of a learner’s misconception and feedback generation. Nevertheless, qualitative data revealed that frequent feedbacks due to the implementation of PM could be obstructive to thought process at advance stage of learning. Thus, PM implementations should also include delayed diagnosis, especially for advance learners who prefer to have it on request. Despite that, current implementation allows users to turn PM off, thereby using alternative learning route. Overall, the research revealed that the implementation of interface-based PM (i.e. conversation and cognitive visibility) improved the visibility of learner’s cognitive process, and this in turn enhanced learning—as perceived.

Wireless technologies and networks are a part of our daily lives and we are surrounded by a constant stream of wireless signals. Unfortunately, there are a lot of limitations. At Toyota Material Handling Europe, future demands and expectations are raising but the technologies available today are not adapting fast enough. Indoor factory environments, moving network nodes and rapidly changing topologies are demanding situations. In this new Internet of Things day and age, transferring information in these challenging environments, the standard master and slave configuration is not enough. This report looks at the specific challenges establishing a wireless communication link between trucks in an indoor factory and warehouse environment. There are many factors to consider. Antennas, noise, frequency bands, different network technologies, propagation and path loss modeling to mention a few. Antennas and on-the-truck antenna positioning require design choices to be made. If we want to benefit from using high gain antennas, positioning is an important factor. Noise, in the highly congested 2.4 GHz band is a problem, especially considering safety critical applications. The license free ISM frequency bands have all different advantages in range, available communication protocols and amount of other technologies sharing the spectrum. The Wi-Fi, Bluetooth, ZigBee and V2X technologies looked at, tailored to the particular case of a warehouse environment like the Toyota factory, and the final selection relates to potential use case scenarios. Prioritizing, scalability, ad hoc network topologies, low latencies and short connection times together with long range, the new V2X technology building upon the IEEE 802.11p standard stands out. This report evaluates the IEEE 802.11p wireless standard running the ETSI ITS G5 V2X protocol in an indoor factory and warehouse environment. Thanks to Kapsch and their EVK-3300 V2X Evaluation Kit it was possible to evaluate transferring information between nodes, the PSR, and looking at how useful different path loss models are for estimating relative distances between trucks.

International Telemetering Conference Proceedings / October 20-23, 2003 / Riviera Hotel and Convention Center, Las Vegas, Nevada / The IMS(In-vehicle Monitoring Subsystem) of VMS(Vehicle Monitoring System) is the multifunctional and complex integrate embedded system, which sends the data of various in-vehicle devices to MC(Monitoring Center) and accepts commands and schedules from there. Using GPRS platform in this system make it possible for real-time and effective data transmission. This paper proposes some new insights on IMS applied to public traffic, including its software and hardware composition, and its realization method.

International Telemetering Conference Proceedings / October 25-28, 1999 / Riviera Hotel and Convention Center, Las Vegas, Nevada / With the rapid development of urban economy, there are bus increasing, route extending, and shuttle frequency increasing etc. At the same time, road construction is subject to land surface, so traffic jam often occurs. It is a big trouble for life of citizens and problem for economy development. So it needs to be improved as fast as possible. Vehicle monitoring system for public traffic in Beijing can expediently monitor the state of each controlled bus, thereby making perfect management. With the integration of GPS, analog trunked communication and digital map, the old, blinding manage system of public traffic would be changed into advanced, visualized management mode, and several routes are dispatched in one dispatch center at the same time. The system frame and its components are introduced in this paper.