LIDAR Sensor Simulation in Adverse Weather Condition for Driving Assistance Development

This paper was produced for the 2019 NAFEMS World Congress in Quebec Canada

Resource Abstract

We present a new LIDAR modelling for automotive applications. Many simulators already exist for this problem, but most of these simulators represent only a punctual modelling of the LASER impacts or cannot simulate the real-time LIDAR operation. The model considers the LASER beam propagation as well as the attenuation of LASER energy not only in clear weather, but also under harsh weather conditions such as fog and rain. The main objective from a LIDAR sensor point of view is to detect all the targets within the observation zone and simultaneously estimate the distance and relative reflectivity of the target in real time. The use of this sensor simulation can easily and effectively replace real LIDAR test campaigns to measure their performance. By comparing real sensors with our model, we demonstrate that this virtual sensor accurately reproduces real LIDAR sensors’ behaviour.