mmCooper: A Multi-agent Multi-stage Communication-efficient and Collaboration-robust Cooperative Perception Framework
This work addresses communication and robustness issues in cooperative perception for autonomous vehicles, representing an incremental improvement over existing methods.
The paper tackled the challenges of bandwidth constraints and calibration errors in cooperative perception for autonomous vehicles by proposing mmCooper, a multi-agent multi-stage framework that dynamically balances information sharing to enhance perception while maintaining communication efficiency, achieving improved accuracy as demonstrated on real-world and simulated datasets.
Collaborative perception significantly enhances individual vehicle perception performance through the exchange of sensory information among agents. However, real-world deployment faces challenges due to bandwidth constraints and inevitable calibration errors during information exchange. To address these issues, we propose mmCooper, a novel multi-agent, multi-stage, communication-efficient, and collaboration-robust cooperative perception framework. Our framework leverages a multi-stage collaboration strategy that dynamically and adaptively balances intermediate- and late-stage information to share among agents, enhancing perceptual performance while maintaining communication efficiency. To support robust collaboration despite potential misalignments and calibration errors, our framework prevents misleading low-confidence sensing information from transmission and refines the received detection results from collaborators to improve accuracy. The extensive evaluation results on both real-world and simulated datasets demonstrate the effectiveness of the mmCooper framework and its components.