Fusion de données pour l'estimation des temps de parcours via la théorie de l'évidence

The emergence of new information technologies and the transformation that has occurred in traffic management have both increased drivers' already considerable need for road traffic information. The travel time is one of the forms in which this information is presented, and a number of systems are based on its dissemination. In this context, this indicator is used as a measure of the impedance (or cost) of routes on the network and/or a congestion indicator. This raises the problem of estimating travel times with an acceptable degree of accuracy, which is a particularly difficult task in urban areas as a result of difficultes of a theoretical, technical and methodological nature. Thus, in order to find out the traffic conditions that prevail on an urban road, the traffic sensors that are usually used to measure traffic conditions are ineffective under certain circumstances. New measurement devices (cameras, GPS or cellphone tracking, etc.) mean that other sources of data are increasingly used in order to supplement the information provided by conventional measurement techniques and improve the accuracy of travel) time estimates. As a result, travel time estimation becomes a typical data fusion problem. This study deals with a multisource estimate of journey times and attempts to provide a comprehensive framework for the utilization of multiple data and demonstrate the feasibility of a travel time estimation system based on the fusion of data of several different types. In this case two types of data are involved, data from conventional induction loop sensors (essentially flow and occupancy measurements) and data from probe vehicles. The selected modelling framework is the Dempster-Shafer Evidence Theory, which has the advantage of being able to take account of both the imprecision and uncertainty of the data. The implementation of this methodology has demonstrated that, in each case, better results are achieved with fusion than with methods based on a single source of data and that the quality of the information, as measured by correctly classified rates, improves as the degree of precision required of the estimate is increased.     

L'influence de la biactivité sur les stratégies spatiales des ménages

A survey conducted among staff at the University of Paris xii Val-de-Marne has shown that there are considerable differences in residential location and commuting strategies between those in one income households and those in two income households. This difference varies depending on social group. In the case of higher social groups, being in a two income household as opposed to a one income household increases the range of residential possibilities rather than the household's income per consumer unit, and in particular allows them to live nearer their work. In contrast, for low income groups, working couples commute considerably greater distances than persons in one income households. Thus, for lower income groups, belonging to a working couple mainly provides greater opportunities for home ownership, as an increase in distances to work does not affect journey times.     

La surestimation de la distance intervéhiculaire dans le brouillard

The difficulties of driving in fog are usually perceived in terms of lower sight distances and reduced ability to anticipate. However, the fall in the quality of visual information can also lead to errors in judgement by affecting the way drivers perceive their environment and their movement within it. We have conducted a series of experiments in a fog chamber and on a driving simulator, with the main aim of investigating the influence of fog on vehicle headway perception. Our results have revealed that distances are seriously overestimated in fog, and that this is likely to contribute to lower headways under such conditions, In fog an overestimation of headway between 20 and 55% was observed when vehicles were visible only as a result of their rear lights, which is the case in night-time or thick daytime fog. Under these conditions, the characteristics of vehicle lights become extremely important. Overestimation of distance becomes greater when vehicles have only one rear light instead of two, and when their two reac lights are close together. Perception of vehicle distance is therefore improved by two rear lights, placed as far apart as possible to indicate the dimensions of the vehicle. No significant overestimation of headways was found during thin or moderate daytime fog, when the vehicle outline remained visible. The results have shown the importance of the familiar size of vehicles as an indicator of distance in the situations we have investigated. They confirm the hypothesis that basic mechanisms of space perception are modified in fog.     

Les objectifs de qualité urbaine et de sécurité routière dans la planification de la mobilit'e plan urbain de trafic (Brescia) et plan de déplacements urbains (Aix-en-Provence), deux visions différentes

Apart from the fact that it is of interest in itself to conduct the first comparison between Urban Travel Plans in France and Italy, this paper has the more general aim of examining a number of ideas which are shared by the scientific community, namely that urban policies and planning practices take ittle account of, or even completely ignore, road safety, the relationship between town planning and transport planning and consultation.     

Utilisation de l'intelligence artificielle distribuée pour la simulation microscopique d'un carrefour

This paper describes an original reasoning mechanism based on the principles of Distributed Artificial Intelligence (PAI). The purpose of this research was to construct a microscopic simulator of the road traffic near an intersection. In this model vehicles are considered as agents whose knowledge base is constructed dynamically before each decision is taken. Knowledge is represented using the production rules formalism. Each element in the environment has conduct rules which are transmitted to all the vehicles which enter its sphere of action. The knowledge base of a vehicle is therefore the result of a concatenation and hierarchization of conduct rules for solely those elements which can influence its bejaviour. A masking procedure is used to implement the priorities which are defined by the scheduling of knowledge and allow the vehicle to have reflex responses. Although we have used just a few easily implemented conduct rules the first results from the simulator have achieved a good standard of realism. This research programme has created an interface which can be used to implement rules and configure and control the simulator.     

Mesure automatique de variables de trafic par traitement d'images application à la régulation en milieu urbain

Inductive loops are widely used for urban traffic control purposes. However, the data they provide cover too small an area to provide an accurate picture of traffic conditions at an intersection. As a result there is increasing interest in image processing techniques as these can provide more varied and more accurate data about the scene whatever the prevalling traffic, environmental and lighting conditions. This paper describes the queue length and occupancy measurement system which we have developed. We begin by outlining the constraints which are imposed by the market and the system's environment. The architecture of the system and the algorithm it uses has been specified with reference to these constraints. An account is then give of the experiments we have conducted in order to evaluate the performance of the system and an analysis of the results is presented.     

Quelle place pour la sécurité routière dans les plans de déplacements urbains? Entre intégration et diffusion,le cas du plan de déplacements urbains de Lille métropole

This paper describes the position given to road safety in Urban Travel Plans by analyzing Urban Travel Plan document from eleven cities and undertakes a more in-depth investigation of road safety in the Urban Community of Lille. These plans provide an opportunity to disseminate knowledge about road safety to the other areas of urban management. They also describe the beginning of an integration of road safety in urban planning, even if intersectoral approaches at city level have yet to be developed and even if, in the case of Lille, the division of institutional competence with regard to the management of public space still requires clarification.     

Évaluation des potentiels de réduction des trafics automobiles dans le cas de la zone dense francilienne

The Pari 21 research project does not attempt to justify a move towards a radically different system—practically everything there is to say about the benefits and disadvantages of cars in cities has already been said. Instead, the project will perform simulations and assess, from the point of view of individuals, the feasibility of a transport system that is based mainly On the use of modes Other than the private car in the densely populated zone Of the Greater Paris Area (Paris city plus the inner suburbs). By constructing and evaluating transport policy scenarios, this research also aims to measure the potential for reducing car use in highly urbanized zones, particularly that which results from the reduction in journey speeds which are a likely consequence of the policies in question The method is based on repeated iterations of a simulation model which combines the assignment of round trips to modes other than the automobile with reformulations of transport supply. Demand is channelled towards personal modes (walking, cycling), public transport and a combination of personal and public modes. These iterations are based on the 1991-1992 Paris Region comprehensive travel survey which recorded all the round trips by car made in a typical day by the individuals living in the zone (i.e. the sequence of all trips made between each time the individual leaves home and returns home).     

Demande de transport de personnes: une théorie unifiée de l'urbain à l'intercontinental Seconde partie: population de l'ensemble des voyages réalisés

This paper proposes a theory which links some of the characteristics of trips made by individuals to the travel price and time provided by transport supply, with reference to the socio-economic context. This model appears to be able to reproduce the phenomena that are observed when there is a change in the supply or context on a given origin-destination link. However, above all, the same model can also reproduce those phenomena (changes in urban development aside) which are different in an urban and an interurban context: greater personal mobility, shorter period away from home, lower value of time among travellers, different modal split for trips, slower increase in the number of trips as a result of the socio-economic context, etc. The first part of the paper shows how each of an individual's potential journeys may be characterized by wishes concerning different aspects of the journey and how the different alternatives for the journey in a given context of transport and accommodation supply can be assessed with reference to the gap between wishes and real characteristics. The second part shows how it is possible to determine the total population of journeys made, with reference to both the context and supply. The theory has been illustrated by numerical results obtained for a variety of situations which are representative of urban and interurban areas.     

Réflexions sur le transfert méthodologique de l’analyse qualitative d’accidents de la circulation routière issue de l’étude détaillée des accidents (EDA) fran?aise aux procédures d’accidents colombiennes

Nearly 90% of the traffic accidents that occur worldwide happen in developing countries. Yet road safety remains a little-studied subject in these countries, and notably accidentology studies, which provide a wealth of information. One of the major problems in studying road accidents in developing countries lies in the lack of reliable accident data. Indeed, while the comprehensive accident analysis methodologies used in developed countries provide an understanding of the origin of accidents and accident processes, as well as making it possible to define suitable, effective actions, they require reliable and relatively detailed accident data. These methodologies are notably based on detailed accident studies (DASs) and their adaptation to the analysis of accident reports. Colombia has an infrastructure for gathering and storing traffic accident data that can be used to perform comprehensive accident analyses. This article defines the feasibility of a technology transfer (clinical accident analysis) to the Colombian equivalent of accident reports. These observations are part of a doctoral thesis prepared jointly by the IFSTTAR Accident Mechanisms Research Unit and the University of the Andes in Bogotá.     

Demande de transport de personnes: une théorie unifiée de l'urbain à l'intercontinental Première partie: modalités de réalisation d'un voyage potentiel

This paper proposes a theory which links some of the characteristics of trips made by individuals to the travel price and time provided by transport supply, with reference to the socio-economic context. This model appears to be able to reproduce the phenomena that are observed when there is a change in the supply or context on a given origin-destination link. However, above all, the same model can also reproduce those phenomena (changes in urban development aside), which are different in an urban and an interurban context: greater personal mobility, shorter period away from home, lower value of time among travellers, different modal split for trips, slower increase in the number of trips as a result of the socio-economic context, etc. The first part of the paper shows how each of an individuals potential journeys may be characterized by wishes concerning different aspects of the journey and how the different alternatives for the journey in a given context of transport and accommodation supply can be assessed with reference to the gap between wishes and real characteristics. The second part shows how it is possible to determine the total population of journeys made with reference to both the context and supply. The theory has been illustrated by numerical results obtained for a variety of situations which are representative of urban and interurban areas.     

Un scénario périurbain de route automatisée

A scenario for a suburban automated highway with the objective of reducing congestion is described. Starting from a reference situation similar to a current traffic lane, the authors consider the influence of a number of parameters on the achievable capacity of an automated section of road. These parameters are used to estimate safe following distances which are computed with the aim of avoiding collisions in the event of sudden braking. They then consider the number of collisions in the presence of a brick wall type obstacle on the carriageway. Ramp metering based on a macroscopic approach leads to a loss of capacity, which we overcome by implementing a management technique that integrates individual vehicles in the stream. Lastly, a general architecture for the scenario with several control and co-ordination layers is proposed.