Presentation Abstract: Cooperative perception is the key approach to augment the perception of ego vehicle toward safe autonomous driving. However, it is challenging to achieve real-time perception sharing for hundreds of vehicles in large-scale deployment scenarios. In this talk, we introduce AdaMap, a new high-scalable real-time cooperative perception system, which achieves guaranteed end-to-end latency under time-varying network dynamics. To achieve AdaMap, we design a tightly coupled data plane and control plane. In the data plane, we design a new hybrid localization module to dynamically switch between object detection and tracking, and a novel encoder-decoder representation module to adaptively compress and recover the point cloud of detected objects. In the control plane, we design a new graph-based object selection method to un-select excessive multi-viewed point clouds of objects, and a novel approximated gradient descent algorithm to optimize the representation factor of point clouds.

Friday, September 15, 2023
11:00 - 11:50 AM Central Time
Nebraska Hall (NH) Room 404 (in person), Lincoln
Peter Kiewit Institute (PKI) Room 160 (remote), Omaha