Learning to Communicate and Correct Pose Errors

Nicholas Vadivelu, Mengye Ren, James Tu, Jingkang Wang, Raquel Urtasun

Research output: Contribution to journalConference articlepeer-review

Abstract

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 [1], 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.

Original languageEnglish (US)
Pages (from-to)1195-1210
Number of pages16
JournalProceedings of Machine Learning Research
Volume155
StatePublished - 2020
Event4th Conference on Robot Learning, CoRL 2020 - Virtual, Online, United States
Duration: Nov 16 2020Nov 18 2020

Keywords

  • multi-agent
  • perception
  • prediction
  • self-driving

ASJC Scopus subject areas

  • Artificial Intelligence
  • Software
  • Control and Systems Engineering
  • Statistics and Probability

Fingerprint

Dive into the research topics of 'Learning to Communicate and Correct Pose Errors'. Together they form a unique fingerprint.

Cite this