Stochastic Geometry Analysis of Collaborative Vision-Based Environment Perception with Blockage Effects

Collaborative vision-based environmental perception plays a vital role in autonomous driving within intelligent transportation systems. By enabling vehicular nodes to exchange local visual data, this approach mitigates occlusions caused by obstacles in the environment. This paper presents a novel analytical framework for collaborative vision-based perception in vehicular networks by applying tools from stochastic geometry and random shape theory. In particular, the road network is modeled as a Poisson line process (PLP), while occlusions from buildings and vehicles are represented as two independent stochastic rectangle processes. Under this setup, the perception probability of an individual vehicle is first derived. Then, we characterize the mean area fraction of environment perception within the circle of radius ${\mathcal{D}}_{p}$ centered at the typical vehicle to evaluate system performance. Simulation results validate our analysis. To the best of our knowledge, this work is the first to conduct a system-level analysis of collaborative vision-based perception in vehicular networks while incorporating random blockages from buildings and vehicles.