Abstract
The robot which can be active outdoors is demanded with a labor shortage and a declining birthrate and aging in recent years. A map of surrounding environment is necessary for autonomous run. As an acquisition system of a map, aline and align a filler the data which acquired during a run one by one, and use iterating overlappart estimation against the problem until now. It is important to a-line at first in using of iterating overlap-part estimation. In this paper, indicate we consider and suggest about accuracy progress in align a filler of scan matching, and adapt first align a filler of global map acquisition method by iterating overlap-part estimation. Align a filler in form comparison gives an error of usually territory in each data. Therefore we plan accuracy progress to extract only characteristic territory and do align a filler. Through we consider of characteristic point extraction.
