Learning to Communicate and Correct Pose Errors
This addresses robustness issues in cooperative perception for self-driving vehicles, though it is incremental as it builds on prior work like V2VNet.
The paper tackles the problem of multi-agent self-driving perception and motion forecasting being vulnerable to pose noise in communication, and proposes a neural reasoning framework that learns to communicate, estimate errors, and reach consensus, significantly improving robustness under realistic noise.
Learned communication makes multi-agent systems more effective by aggregating distributed information. However, it also exposes individual agents to the threat of erroneous messages they might receive. In this paper, we study the setting proposed in V2VNet, where nearby self-driving vehicles jointly perform object detection and motion forecasting in a cooperative manner. Despite a huge performance boost when the agents solve the task together, the gain is quickly diminished in the presence of pose noise since the communication relies on spatial transformations. Hence, we propose a novel neural reasoning framework that learns to communicate, to estimate potential errors, and finally, to reach a consensus about those errors. Experiments confirm that our proposed framework significantly improves the robustness of multi-agent self-driving perception and motion forecasting systems under realistic and severe localization noise.