Attention Graph for Multi-Robot Social Navigation with Deep Reinforcement Learning

In this paper, we present MultiSoc, a new method for learning multi-agent socially aware navigation strategies using RL. Inspired by recent works on multi-agent deep RL, our method leverages graph-based representation of agent interactions, combining the positions and fields of view of entities (pedestrians and agents). Each agent uses a model based on two Graph Neural Networks combined with attention mechanisms. First an edge-selector produces a sparse graph, then a crowd coordinator applies node attention to produce a graph representing the influence of each entity on the others. This is incorporated into a model-free RL framework to learn multi-agent policies. We evaluate our approach on simulation and provide a series of experiments in a set of various conditions that conclude that our method learns faster than social navigation deep RL mono-agent techniques, and enables efficient multi-agent implicit coordination in challenging crowd navigation with multiple heterogeneous humans. A full description and analysis of this work is available in the full paper version [2].