Environment map generation in forest using field robot

Abstract

The employment rate of young people in Japanese forestry tends to decline, and the unmanaged forest area is expected to increase in the future. Therefore, in our laboratory we propose an autonomous field robot with all terrain vehicles that focuses on the automation of work. The robot automates weeding and observation in the forest. In this research, we observed trees by generating an environmental map in the forest using Simultaneous Localization and Mapping (SLAM). The error of the generated environmental map was about 1[m].