Abstract: The performance of LIDAR sensors in autonomous vehicles is critical to ensuring safety and reliability, especially under different weather conditions. In this paper, we present a simulation that evaluates the performance of LIDAR sensors under sunny, rainy, and foggy weather conditions using the High-Fidelity Weather Model (HFWXM). The simulation aimed to assess the LIDAR sensor ‘s ability to detect and track objects accurately under different weather conditions. The results of the simulation showcased the considerable impact of the prevailing weather conditions on the performance of the LIDAR sensor. Notably, the detection rate exhibited an average value of 95.2%, 82.5%, and 71.8% during the sunny, rainy, and foggy atmospheres, respectively. The implications of such divergent weather patterns in their resulting false positive rates revealed themselves as having a higher magnitude under the muddied and misty domains (2.5% and 4.7%, respectively), in contrast to the sunny weather ‘s more modest rates (1.2%). The tracking accuracy, too, exhibited a marked contrast between the weather conditions, with the tracking accuracy dipping under the rainy (0.14 m) and foggy (0.23 m) climates when compared with the sun-soaked and shining environment (0.11 m). The simulation findings highlight the need to consider the LIDAR sensor ‘s performance under different weather conditions while designing autonomous vehicles to ensure their safety and reliability. ORCID NO: 0000-0003-4661-7957, 0000-0001-8972-1952
Anahtar Kelimeler: LIDAR sensors, Autonomous vehicles, Weather conditions, Perception
|
