Performance of drivable path detection system of autonomous robots in rain and snow scenario

Abstract

Drivable path detection is an important factor to consider for a successful development of autonomous robot which is characterized as an intelligent vehicle. Researchers using different vision-based techniques have achieved remarkable result toward drivable path detection. Regardless of this achievement, environmental noise such as rain and/or snow can cause misdetection of drivable path which can lead to autonomous robot accident. In this paper, after investigating the effects of rain and/or snow, we introduced into the drivable path detection system a filtering algorithm that addresses the detection and removal of rain and/or snow for the optimization of the system. Experiments were carried out to show the effectiveness of the filter in the system. The results show that filtering algorithm assists the autonomous driving system in navigating perfectly during rain and/or snow scenario with minimal accident.