Multi-Robot Collaborative Perception with Graph Neural Networks

TL;DR

This paper introduces a graph neural network framework to improve perception accuracy and robustness in multi-robot systems, especially under noisy or occluded conditions, by enabling effective perception sharing and fusion among robots.

Contribution

The paper presents a novel GNN-based approach for multi-robot perception that enhances inference accuracy and resilience to sensor failures, applicable to tasks like depth estimation and segmentation.

Findings

Improved perception accuracy in noisy and occluded conditions.

Enhanced resilience to sensor failures and disturbances.

Effective multi-view perception fusion demonstrated on aerial robot data.

Abstract

Multi-robot systems such as swarms of aerial robots are naturally suited to offer additional flexibility, resilience, and robustness in several tasks compared to a single robot by enabling cooperation among the agents. To enhance the autonomous robot decision-making process and situational awareness, multi-robot systems have to coordinate their perception capabilities to collect, share, and fuse environment information among the agents in an efficient and meaningful way such to accurately obtain context-appropriate information or gain resilience to sensor noise or failures. In this paper, we propose a general-purpose Graph Neural Network (GNN) with the main goal to increase, in multi-robot perception tasks, single robots' inference perception accuracy as well as resilience to sensor failures and disturbances. We show that the proposed framework can address multi-view visual perception problems such as monocular depth estimation and semantic segmentation. Several experiments both using photo-realistic and real data gathered from multiple aerial robots' viewpoints show the effectiveness of the proposed approach in challenging inference conditions including images corrupted by heavy noise and camera occlusions or failures.