Experimental evaluation of a method for simulation based learning for a multi-agent system acting in a physical environment

Publikation: Kapitel i bog/rapport/konference-proceedingKonferencebidrag i proceedingsForskningpeer review

128 Downloads (Pure)

Abstrakt

A method for simulation based reinforcement learning (RL) for a multi-agent system acting in a physical environment is introduced, which is based on Multi-Agent Actor-Critic (MAAC) reinforcement learning. In the proposed method, avatar agents learn in a simulated model of the physical environment and the learned experience is then used by agents in the actual physical environment. The proposed concept is verified using a laboratory benchmark setup in which multiple agents, acting within the same environment, are required to coordinate their movement actions to prevent collisions. Three state-of-the-art algorithms for multi-agent reinforcement learning (MARL) are evaluated, with respect to their applicability for a predefined benchmark scenario. Based on simulations it is shown that the MAAC method is most applicable for implementation as it provides effective distributed learning and suits well to the concept of learning in simulated environments. Our experimental results, which compare simulated learning and task execution in a simulated environment with that of task execution in a physical environment demonstrate the feasibility of the proposed concept.

OriginalsprogEngelsk
TitelProceedings of the 11th International Conference on Agents and Artificial Intelligence
RedaktørerAna Rocha, Luc Steels, Jaap van den Herik
Vol/bind1: ICAART
ForlagSCITEPRESS Digital Library
Publikationsdato2019
Sider103-109
ISBN (Elektronisk)9789897583506
DOI
StatusUdgivet - 2019
Begivenhed11th International Conference on Agents and Artificial Intelligence, ICAART 2019 - Prague, Tjekkiet
Varighed: 19. feb. 201921. feb. 2019

Konference

Konference11th International Conference on Agents and Artificial Intelligence, ICAART 2019
LandTjekkiet
ByPrague
Periode19/02/201921/02/2019
SponsorInstitute for Systems and Technologies of Information, Control and Communication (INSTICC)

Fingeraftryk

Dyk ned i forskningsemnerne om 'Experimental evaluation of a method for simulation based learning for a multi-agent system acting in a physical environment'. Sammen danner de et unikt fingeraftryk.

Citationsformater