Multi-Robot Collaborative Perception with Graph Neural Networks
This work addresses the need for robust and efficient collaborative perception in multi-robot systems like aerial swarms, offering incremental improvements in inference accuracy and resilience.
The paper tackles the problem of enhancing perception accuracy and resilience in multi-robot systems by proposing a Graph Neural Network framework, achieving improved performance in tasks like monocular depth estimation and semantic segmentation under challenging conditions such as heavy noise and camera failures.
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.