Abstract
In this paper, We describe our research aiming to automatically detect damaged sections of sidewalk using SOKUIKI sensor mounted on a mobile robot. We consider the problem of sidewalk survey and purpose of our research to solve that. First, we consider sensing method using a mobile robot equiped with UTM-30LX, which can get three demention value. Next we consider identification method of damaged data competing all data within a certain range with some reference planes. They are made by three points of grid data that are extracted to random in the range of conditions. Finaly, we verificate these devised mothods to apply five patterns of sidewalk. In the result, all of them can be identificated damaged sections that are similer to result of the goal, so we think that devised methods are varid to detect damaged sections of sidewalk.
