A Scalable and Parallelizable Digital Twin Framework for Sustainable Sim2Real Transition of Multi-Agent Reinforcement Learning Systems

arXiv:2403.10996v1 Announce Type: cross
Abstract: This work presents a sustainable multi-agent deep reinforcement learning framework capable of selectively scaling parallelized training workloads on-demand, and transferring the trained policies from simulation to reality using minimal hardware resources. We introduce AutoDRIVE Ecosystem as an enabling digital twin framework to train, deploy, and transfer cooperative as well as competitive multi-agent reinforcement learning policies from simulation to reality. Particularly, we first investigate an intersection traversal problem of 4 cooperative vehicles (Nigel) that share limited …

abstract agent arxiv cs.lg cs.ma cs.ro demand digital digital twin ecosystem enabling framework hardware learning systems multi-agent reality reinforcement reinforcement learning resources scalable scaling simulation sustainable systems training transition twin type work workloads