{"id":284,"date":"2022-12-07T16:04:08","date_gmt":"2022-12-07T15:04:08","guid":{"rendered":"https:\/\/www.mines-stetienne.fr\/fayol\/?page_id=284"},"modified":"2022-12-09T14:24:36","modified_gmt":"2022-12-09T13:24:36","slug":"these-preparee-par-alaa-daoud","status":"publish","type":"page","link":"https:\/\/www.mines-stetienne.fr\/fayol\/theses\/these-preparee-par-alaa-daoud\/","title":{"rendered":"Th\u00e8se pr\u00e9par\u00e9e par Alaa Daoud"},"content":{"rendered":"\n\nTitre : Approche d\u00e9centralis\u00e9e pour l’allocation de courses \u00e0 la demande \u00e0 une flotte de v\u00e9hicules autonomes<\/strong>\n\nTitle : Decentralized On-Demand Resource Allocation for Autonomous Vehicle Fleets<\/strong>\n\nBeginning of thesis\u00a0:<\/strong> 2018\nEnd of thesis\u00a0:<\/strong> 2021\n\nR\u00e9sum\u00e9 : <\/strong>Le d\u00e9veloppement de v\u00e9hicules autonomes, capables de communiquer de pair \u00e0 pair, ainsi que l’int\u00e9r\u00eat pour les solutions \u00e0 la demande sont les principales motivations de cette \u00e9tude. En l’absence de contr\u00f4le central, nous sommes int\u00e9ress\u00e9s par la formation d’une flotte de v\u00e9hicules autonomes capables de r\u00e9pondre aux demandes de d\u00e9placement \u00e0 l’\u00e9chelle de la ville.\nTypiquement, ce probl\u00e8me est r\u00e9solu de mani\u00e8re centralis\u00e9e, ce qui implique que les v\u00e9hicules ont un acc\u00e8s continu \u00e0 un portail de r\u00e9partition. Cependant, cet acc\u00e8s \u00e0 une telle infrastructure de commutation mondiale (pour la collecte de donn\u00e9es et la livraison des commandes) est co\u00fbteux et repr\u00e9sente un goulot d’\u00e9tranglement critique. L’id\u00e9e est d’utiliser des technologies de communication de v\u00e9hicule \u00e0 v\u00e9hicule (V2V) \u00e0 faible co\u00fbt pour coordonner les v\u00e9hicules sans infrastructure de communication globale.\nNous proposons de mod\u00e9liser les diff\u00e9rents aspects des probl\u00e8mes de d\u00e9cision et d’optimisation li\u00e9s \u00e0 ce probl\u00e8me plus g\u00e9n\u00e9ral. Apr\u00e8s avoir mod\u00e9lis\u00e9 ces probl\u00e8mes, la question se pose du choix des m\u00e9thodes de solution centralis\u00e9es et d\u00e9centralis\u00e9es. Sur le plan m\u00e9thodologique, nous explorons les orientations et comparons les performances des techniques d’optimisation des contraintes distribu\u00e9es (DCOP), des techniques multi-agents auto-organis\u00e9es, des approches bas\u00e9es sur le march\u00e9 et des solutions centralis\u00e9es de recherche op\u00e9rationnelle.\u00a0\n<\/strong>\n\nAbstract : <\/strong>The development of autonomous vehicles, capable of peer-to-peer communication, as well as the interest in on-demand solutions are the primary motivations for this study. In the absence of central control, we are interested in forming a fleet of autonomous vehicles capable of responding to city-scale travel demands.\nTypically, this problem is solved centrally, this implies that the vehicles have continuous access to a dispatching portal. However, such access to such a global switching infrastructure (for data collection and order delivery) is costly and represents a critical bottleneck. The idea is to use low-cost vehicle-to-vehicle (V2V) communication technologies to coordinate vehicles without a global communication infrastructure.\nWe propose to model the different aspects of decision and optimization problems related to this more general problem. After modeling these problems, the question arises as to the choice of centralized and decentralized solution methods. Methodologically, we explore the directions and compare the performance of distributed constraint optimization techniques (DCOP), self-organized multi-agent techniques, market-based approaches, and centralized operations research solutions.\n\nMots cl\u00e9s\u00a0:<\/strong> Transport \u00e0 la demande, V\u00e9hicules autonomes, Allocation des ressources, Syst\u00e8mes multi-agents\n\nKeywords<\/strong> : Multi-agent systems, On-demand transport, Resource allocation, Autonomous vehicles\n\nDate de Soutenance pr\u00e9vue : <\/strong> dernier trimestre de 2021\n\nEncadrement :<\/strong>\n