Approches multiagents pour l’allocation de courses à une flotte de taxis autonomes

Abstract : This work is interested in decentrally solving a taxi allocation problem over a fleet of autonomous taxis. Classically, to solve this problem, requests are centralized into a portal where a dispatcher allocates requests to taxis (ideally, in an optimal manner). This requires taxis have continuous access to the portal. However, getting access to such global communication infrastructureis very expensive for taxi companies. The idea here is to use new affordable vehicle-to-vehicle communication technologies to coordinate taxis without global communication infrastructure. Our approach is presented and empirically evaluated via simulations. We have developed different scenarios with different communication infrastructure and coordination mechanisms, and we analyze, their resulting quality of service, user welfare, gain and robustness to message loss.
Document type :
Journal articles
Complete list of metadatas

https://hal-emse.ccsd.cnrs.fr/emse-02303732
Contributor : Florent Breuil <>
Submitted on : Wednesday, October 2, 2019 - 3:05:24 PM
Last modification on : Tuesday, November 19, 2019 - 2:37:48 AM

Identifiers

Citation

Gauthier Picard, Flavien Balbo, Olivier Boissier. Approches multiagents pour l’allocation de courses à une flotte de taxis autonomes. Revue des Sciences et Technologies de l'Information - Série RIA : Revue d'Intelligence Artificielle, Lavoisier, 2018, 32 (2), pp.223-247. ⟨10.3166/RIA.32.223-247⟩. ⟨emse-02303732⟩

Share

Metrics

Record views

54