Map refinement for mobile robot navigation by polygonal approximation and distance transformation
Deaho Lee, Sung-Soo Lim, Younjung Lee, Jaeyoung Yoon · 2016
We propose a refinement method of maps built by optical sensors for mobile robot navigation. The original map built by optical sensors has much noisy information; however, the map may lose significant information when the maps are refined by common noise-reduction filters, therefore the original maps are generally used for robot navigation. To minimize lost information and remove boundary noises, we use morphological operators and polygonal approximation for level regions, and use distance transform to preserve non-ground regions which contact with level regions in this paper. The experimental results show that the proposed method can efficiently remove boundary noises and can preserve significant information. Therefore, the method can be used for mobile robot navigation instead of original built maps.