Apprentissage constructiviste à base de systèmes multiagents. Une application au problème complexe de la régulation coopérative du trafic

Maxime Guériau 1, 2 Frédéric Armetta 1 Salima Hassas 1 Romain Billot 3, 4 Nour-Eddin El Faouzi 2
1 SMA - Systèmes Multi-Agents
LIRIS - Laboratoire d'InfoRmatique en Image et Systèmes d'information
3 Lab-STICC_IMTA_CID_DECIDE
Lab-STICC - Laboratoire des sciences et techniques de l'information, de la communication et de la connaissance
Abstract : Decision making in autonomous systems is particularly challenging in unknown and changing complex environments, where providing a complete a priori representation is not possible. The so built representation should be the result of the system interactions with the environment. To illustrate the problem, we consider a decentralized control of road traffic, where a control device of the distributed infrastructure locally controls traffic by sending recommendation messages to connected vehicles. We propose an approach able to combine, without prior domain-knowledge, a set of existing traditional unsupervised learning methods that collaborate as a population of agents in order to build an efficient representation. This study addresses the main scientific issues to consider for such a system to efficiently learn. Our approach follows a constructivist learning perspective, where a population of agents is able to collectively build a representation that dynamically combines discretization processes.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01885654
Contributor : Frédéric Armetta <>
Submitted on : Tuesday, October 2, 2018 - 10:56:58 AM
Last modification on : Tuesday, February 26, 2019 - 11:36:02 AM

Identifiers

Citation

Maxime Guériau, Frédéric Armetta, Salima Hassas, Romain Billot, Nour-Eddin El Faouzi. Apprentissage constructiviste à base de systèmes multiagents. Une application au problème complexe de la régulation coopérative du trafic. Revue des Sciences et Technologies de l'Information - Série RIA : Revue d'Intelligence Artificielle, Lavoisier, 2018, 32/2, pp.249-277. ⟨10.3166/ria.32.249-277⟩. ⟨hal-01885654⟩

Share

Metrics

Record views

251