Les six articles de ces actes concernent les domaines routiers, des pietons et du transport combi... more Les six articles de ces actes concernent les domaines routiers, des pietons et du transport combine rail-route. La description, la reproduction, la prevision, l'optimisation du fonctionnement des reseaux de transport, font appel a des modelisations, avec des approches sciences humaines (perception, decision, action des pietons) et mathematiques. Bernard Schnetzler, Xavier Louis, Jean-Patrick Lebacque developpent un modele de trafic du second ordre avec des diagrammes fondamentaux parametres en fonction de la longueur des vehicules. Sabine Limbourg optimise le nombre et la localisation des terminaux dedies au transbordement de fret continental pour le transport rail-route transeuropeen. Ioannis Papamichail, Markos Papageorgiou, Vincent Vong et John Gaffney presentent HERO, une strategie de regulation coordonnee d’acces autoroutiers ; ses benefices sont montres sur une application a Monash (Australie). Jean-Michel Auberlet presente l’architecture d’une modelisation des deplacements pietons (perception, niveaux tactique et strategique) et les perspectives de la recherche en montrant les difficultes liees aux donnees, au passage a l’echelle. Victorin Martin, Jean-Marc Lasgouttes et Cyril Furtlehner fondent une prevision probabiliste des temps de parcours sur les donnees de vehicules traceurs, en ponderant celles provenant des sections voisines. Mehdi Keyvan-Ekbatani, Markos Papageorgiou et Ioannis Papamichail developpent une commande optimale des reseaux urbains tenant compte d’un diagramme fondamental de reseau et optimisant les temps de presence des vehicules.
Abstract Microscopic modeling of human driving consists generally in combining both car-following... more Abstract Microscopic modeling of human driving consists generally in combining both car-following and lane-change models. While the human car-following process has been extensively developed and well modeled, the lane-change behavior is more complex to understand and still remains to be explored. Classical lane-change models are usually rule-based and handcrafted, that tend to exhibit limited performance. Machine Learning algorithms, particularly Reinforcement Learning (RL) ones, provide an alternative approach and have recently achieved high success in modeling difficult decision-making processes in many fields. We propose in this article a reinforcement learning based model for the human lane-change behavior, with an online calibration of real lane-change decisions, extracted from the NGSIM data-set. In addition, we use the traffic vehicular simulator SUMO ("Simulation of Urban Mobility") to create a numerical simulation environment. The utilization of numerical traffic simulation allows us enriching the data-set, for training the agent to find an optimal policy for lane change. Thus, about 13% additional traffic situations, not present in the real data, are created by the traffic simulation environment. The trained agent is collision-free and human-like who is satisfactory to real data and also to the additional simulated data. Moreover, our RL model can perform up to 95.37% of the real decisions observed in the data-set.
Traffic control in mass transit consists of the regulation of both vehicle dynamics and passenger... more Traffic control in mass transit consists of the regulation of both vehicle dynamics and passenger flows. While most of the existing approaches focus on the optimization of vehicle dwell time, vehicle time headway, and passenger stocks, we propose in this article an approach which also includes the optimization of the passenger inflows to the platforms. We developed in this work a deep reinforcement Q-learning model for the traffic control in a mass transit line. We first propose a new mathematical traffic model for the train and passengers dynamics. The model combines a discrete-event description of the vehicle dynamics, with a macroscopic model for the passenger flows. We use this new model as the environment of the traffic in mass transit for the reinforcement learning optimization. For this aim, we defined, under the new traffic model, the state variables as well as the control ones, including in particular the number of running vehicles, the vehicle dwell times at stations, and ...
Les six articles de ces actes concernent les domaines routiers, des pietons et du transport combi... more Les six articles de ces actes concernent les domaines routiers, des pietons et du transport combine rail-route. La description, la reproduction, la prevision, l'optimisation du fonctionnement des reseaux de transport, font appel a des modelisations, avec des approches sciences humaines (perception, decision, action des pietons) et mathematiques. Bernard Schnetzler, Xavier Louis, Jean-Patrick Lebacque developpent un modele de trafic du second ordre avec des diagrammes fondamentaux parametres en fonction de la longueur des vehicules. Sabine Limbourg optimise le nombre et la localisation des terminaux dedies au transbordement de fret continental pour le transport rail-route transeuropeen. Ioannis Papamichail, Markos Papageorgiou, Vincent Vong et John Gaffney presentent HERO, une strategie de regulation coordonnee d’acces autoroutiers ; ses benefices sont montres sur une application a Monash (Australie). Jean-Michel Auberlet presente l’architecture d’une modelisation des deplacements pietons (perception, niveaux tactique et strategique) et les perspectives de la recherche en montrant les difficultes liees aux donnees, au passage a l’echelle. Victorin Martin, Jean-Marc Lasgouttes et Cyril Furtlehner fondent une prevision probabiliste des temps de parcours sur les donnees de vehicules traceurs, en ponderant celles provenant des sections voisines. Mehdi Keyvan-Ekbatani, Markos Papageorgiou et Ioannis Papamichail developpent une commande optimale des reseaux urbains tenant compte d’un diagramme fondamental de reseau et optimisant les temps de presence des vehicules.
Abstract Microscopic modeling of human driving consists generally in combining both car-following... more Abstract Microscopic modeling of human driving consists generally in combining both car-following and lane-change models. While the human car-following process has been extensively developed and well modeled, the lane-change behavior is more complex to understand and still remains to be explored. Classical lane-change models are usually rule-based and handcrafted, that tend to exhibit limited performance. Machine Learning algorithms, particularly Reinforcement Learning (RL) ones, provide an alternative approach and have recently achieved high success in modeling difficult decision-making processes in many fields. We propose in this article a reinforcement learning based model for the human lane-change behavior, with an online calibration of real lane-change decisions, extracted from the NGSIM data-set. In addition, we use the traffic vehicular simulator SUMO ("Simulation of Urban Mobility") to create a numerical simulation environment. The utilization of numerical traffic simulation allows us enriching the data-set, for training the agent to find an optimal policy for lane change. Thus, about 13% additional traffic situations, not present in the real data, are created by the traffic simulation environment. The trained agent is collision-free and human-like who is satisfactory to real data and also to the additional simulated data. Moreover, our RL model can perform up to 95.37% of the real decisions observed in the data-set.
Traffic control in mass transit consists of the regulation of both vehicle dynamics and passenger... more Traffic control in mass transit consists of the regulation of both vehicle dynamics and passenger flows. While most of the existing approaches focus on the optimization of vehicle dwell time, vehicle time headway, and passenger stocks, we propose in this article an approach which also includes the optimization of the passenger inflows to the platforms. We developed in this work a deep reinforcement Q-learning model for the traffic control in a mass transit line. We first propose a new mathematical traffic model for the train and passengers dynamics. The model combines a discrete-event description of the vehicle dynamics, with a macroscopic model for the passenger flows. We use this new model as the environment of the traffic in mass transit for the reinforcement learning optimization. For this aim, we defined, under the new traffic model, the state variables as well as the control ones, including in particular the number of running vehicles, the vehicle dwell times at stations, and ...
Uploads
Papers by Nadir Farhi