Numeric Analysis of Polarized Lidar Contrast Enhancement Across Diverse Targets in Fog.

Lidar systems are of major importance for driver assistance systems and autonomous driving, but their obstacle detection range can be heavily impaired by adverse weather conditions like fog. To mitigate the effects of fog, the use of polarized light has been proposed. We simulated a polarized lidar system, taking Mie theory-based scattering functions for different fog types at several wavelengths into account, and investigated the possible increase in contrast between fog backscattering and target returns for different surface types, depending on the target orientation and the polarizer configuration. Fog is modeled as an infinite, homogeneous, scattering, and absorbing medium. The simulated detector registers the time-dependent radiance, resolved by scattering order. Our findings show that the suitable choice of illumination and detection polarization allows the amount of single-scattered light detected to be reduced by two to four orders of magnitude and the contrast between fog and target to be increased significantly. Crucially, this approach delivers robust contrast enhancement regardless of whether the target surface is depolarizing Lambertian or perfectly reflecting.