One of today's most serious social, economical and environmental problems is traffic congestion. In addition to the financial cost of the problem, the number of traffic related injuries and casualties is very high. A recently considered approach to increase safety while reducing congestion and improving driving conditions is Automated Highway Systems (AHS). The AHS will evolve from the present highway system to an intelligent vehicle/highway system that will incorporate communication, vehicle control and traffic management techniques to provide safe, fast and more efficient surface transportation. A key factor in AHS deployment is intelligent vehicle control. While the technology to safely maneuver the vehicles exists, the problem of making intelligent decisions to improve a single vehicle's travel time and safety while optimizing the overall traffic flow is still a stumbling block. We propose an artificial intelligence technique called stochastic learning automata to design an intelligent vehicle path controller. Using the information obtained by on-board sensors and local communication modules, two automata are capable of learning the best possible (lateral and longitudinal) actions to avoid collisions. This learning method is capable of adapting to the automata environment resulting from unmodeled physical environment. Simulations for simultaneous lateral and longitudinal control of an autonomous vehicle provide encouraging results. Although the learning approach taken is capable of providing a safe decision, optimization of the overall traffic flow is also possible by studying the interaction of the vehicles. The design of the adaptive vehicle path planner based on local information is then carried onto the interaction of multiple intelligent vehicles. By analyzing the situations consisting of conflicting desired vehicle paths, we extend our design by additional decision structures. The analysis of the situations and the design of the additional structures are made possible by the study of the interacting reward-penalty mechanisms in individual vehicles. The definition of the physical environment of a vehicle as a series of discrete state transitions associated with a "stationary automata environment" is the key to this analysis and to the design of the intelligent vehicle path controller. This work was supported in part by the Center for Transportation Research and Virginia DOT under Smart Road project, by General Motors ITS Fellowship program, and by Naval Research Laboratory under grant no. N000114-93-1-G022. / Ph. D.

Cette thèse est consacrée à la conception et au développement d’une plateforme multi-agent, en réalité virtuelle, de pilotage de véhicules intelligents pour la simulation du comportement microscopique du trafic. D’abord, un système de simulation intelligent des véhicules en réalité virtuelle (VR-ISSV), basé sur les multi-agents, est proposé : c’est un système modulaire hiérarchique de modélisation et de simulation, comprenant une couche matérielle, réseau et système d’exploitation ; une couche de gestion de la visualisation ; une couche de multi-agents et une couche d’interface homme-machine. Ensuite, pour le modèle d’agent du véhicule intelligent, un paradigme de conception décentralisée est utilisé basé sur l’approche multi-contrôleurs, où le comportement du suivi des véhicules et le comportement du dépassement des véhicules sont réalisées par coordination entre multi-contrôleurs. L’agent d’environnement est construit en tenant compte de l’interaction entre les véhicules et l’environnement naturel synthétique. Un système d’information géographique (GIS) est par ailleurs utilisé afin de définir l’agent d’environnement. Enfin, pour assurer la sécurité dans les manœuvres microscopiques du trafic, plusieurs contrôleurs du véhicule intelligent, adaptés à l’environnement complexe, sont considérés. Les contrôleurs, basés sur la logique floue, sont proposés pour envoyer les commandes appropriées aux actionneurs du véhicule - volant de direction, accélérateur, frein... Les modèles de comportement microscopique du trafic basé sur l’agent de véhicule intelligent sont étudiés considérant différents scénarios et l’environnement / This PhD thesis is dedicated to the modeling and simulation of microscopic traffic behavior in virtual reality system, with the intent of providing a new approach to effectively ensure traffic safety. At first, Virtual Reality Intelligent Simulation System of Vehicles (VR-ISSV), based on multi-agent, is proposed to simulate the intelligent microscopic traffic, which is a hierarchical modular modeling and simulation system consisting of hardware, network and operating system layers, visualization management layer, multi-agent layer, human-machine interface layer. The multi-agent layer includes entity agents (intelligent vehicle agents and around vehicle agents), service agent and environment agent. Second, for the intelligent vehicle agent model, a decentralized design paradigm is used for developing the multi-controller based intelligent vehicle, whereby the car following behavior and the overtaking behavior could be realized by the coordination of the multi-controller. The environment agent is constructed based on the conception of Synthetic Natural Environment (SNE), taking into account the interaction between the vehicles and the natural environment. Geographic Information System (GIS) is used to establish environment agent. Finally, to ensure the safety in microscopic traffic maneuver, the intelligent vehicle controllers adapting to complex environment are considered. Fuzzy logic based controllers are designed for sending the appropriate outputs to the vehicle’s actuators – the steering wheel and the throttle/brake pedals. Microscopic traffic behavior models based on the intelligent vehicle agent involving environment are studied