Redundancy Mitigation in Cooperative Perception for Connected and Automated Vehicles

Abstract

Cooperative perception (or cooperative sensing or collective perception) enables connected and automated vehicles to exchange sensor data in order to improve their perception of the driving environment. ETSI is currently developing a standard for collective perception. The standard defines the message format and generation rules. These rules identify when a message should be transmitted and what information it should include. This study shows first that the current ETSI solution generates many redundant collective perception messages that increase the channel load and can compromise the networks' scalability. Unnecessary redundancy can reduce the reliability of V2X (Vehicle to Everything) communications and ultimately decrease the effectiveness of collective perception. This study proposes a modification of the current ETSI solution to control redundancy and avoid the transmission of unnecessary CPM data or messages. The evaluation shows that our proposal significantly reduces the redundancy and channel load and improves the reliability of V2X communications compared to current ETSI solution for collective perception. This is achieved while maintaining the perception achieved by ETSI for the safety-critical short and medium distances.