Collective Perception service for Connected Vehicles and Roadside Infrastructure

Abstract

Collective Perception Service (CPS) is under standardisation at European Telecommunications Standardisation Institute (ETSI) with the objective of enabling sensor-equipped vehicles and roadside units (RSU) to notify nearby vehicles of the objects detected by their sensors. This service is expected to significantly improve perception of vehicles and, hence, improve road safety. While both vehicles and RSUs can contribute in CPS, they have very different characteristics in terms of mobility, sensor field of view (FOV), communication coverage, processing capacity, and service integration cost. This paper compares CPS provided by vehicles and by RSUs, particularly their impacts on extending vehicles’ perception. We will first develop an analytical model that formulates the number of objects perceived by vehicles and RSUs and the success probability of CPM targeting the IEEE 802.11p technology. The analytical results will then be confirmed by simulation evaluations conducted using the VEINS simulator. The analytical and simulation results reveal that, RSU-assisted CPS significantly outperforms vehicle-assisted CPS, providing up to 8-times higher effective number of perceived objects. The results suggest a great opportunity of optimising radio resource utilisation without degrading collective perception by exploiting roadside infrastructure.