.png){kind=icon}
.png){kind=icon}
Return
Physics-informed multi-agent reinforcement learning with multi-scale graph perception for multi-robot coordination
.png){kind=icon}
PRE
AI
H
钟
Z
D
DOI：10.1016/j.eswa.2026.133930.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Physics-guided control improves coordination in large teams of robots. • Long-range team awareness is combined with stable local decisions. • The method improves navigation and spatial sampling performance. • Robust performance is maintained under noisy links and robot failures. • The method scales to larger robot teams without additional training.
Keywords:
Multi-robot systems
Multi-agent reinforcement learning
Physics-informed control
Port-Hamiltonian systems
Graph neural networks
Long-range coordination
Journal
IF:
7.5
Papers:
2.9W
Citations:
10.2W
Organization
No organization information available
Cited Papers