Intensity-Based Identification of Reflective Surfaces for Occupancy Grid Map Modification

The presence of mobile robots in indoor environments is becoming increasingly significant. While there are algorithms such as SLAM that allow to navigate these environments, they do not take into account structural elements modern buildings large glass areas. Due the use light-based sensors laser scans, only visible at few incident angles, so seen noise by algorithms. In this work, reflective surfaces detected analyzing scan intensity values order build a reflection-aware map for safe robot navigation. main novelty proposed method usage single sensor gather environment information and parameter determine where zones are. Experiments carried out multiple complex real scenarios, quantitatively testing how well performs against state art works factors changes orientation, number areas or illumination.