Analysis of Automotive Lidar Sensor Model Considering Scattering Effects in Regional Rain Environments

Abstract

Automotive Lidar sensors are highly susceptible to their environment. One of its major limitations results from the effects of rain environments, which should be seriously considered while designing a Lidar system. This study addresses the impact of rain on the Lidar system by considering the raindrop distributions of different regions. The regional distributions are derived from the rainfall data of three locations, which were reported by previous works, and converted using the constrained-gamma model. The regional distribution reveals different characteristics of raindrops, such as sizes, shapes, and numbers. The derived raindrop distributions are imported to a custom-built Lidar model, providing three models representing the three regions. The simulation results demonstrate that the signal power received by a Lidar attenuates, which is modeled using Mie scattering theory, and the amount of attenuation clearly differs in the regional models. Therefore, the attenuation characteristics change according to the regions; consequently, their effect on the Lidar sensor performances are quantitatively evaluated. In addition, the custom-built Lidar model is mounted on a virtual vehicle, which is simulated using a commercial automobile testing software, PreScan. The driving simulation also demonstrates similar conclusion that the regional raindrop distribution is critical in determining the Lidar performances.