Experimental analysis of raindrop-induced SNR attenuation in MEMS Lidar

Adverse weather significantly degrades Lidar performance in autonomous driving. Atmosphere particles scatter and attenuate the laser beam - altering its direction, divergence, and intensity - which in turn impairs the system detectability. The attenuation for laser intensity directly lowers the signal-to-noise ratio (SNR) of Lidar, and SNR is crucial for reliable detection. This paper performed a three-part investigation. Firstly, we explore how SNR varies with detecting range a controlled indoor environment without extraneous disturbances. Secondly, with the same Lidar emission setup, we measured the echo waveform and SNR at a single point under moderate rainfall to quantify the impact of raindrops. Finally, we derived and present a fitting curve that models the relationship between SNR and detection distance in moderate rain. The result would give some advices for Lidar application in complex weather.