主催: 一般社団法人 日本機械学会
会議名: ロボティクス・メカトロニクス 講演会2023
開催日: 2023/06/28 - 2023/07/01
In Japan, the number of infrastructure facilities to be inspected is rapidly increasing. Currently, infrastructure facilities are inspected manually, but the number of inspection workers is decreasing. Therefore, the introduction of robots is required to improve efficiency and reduce manpower. sual odometry is expected to provide highly accurate self-position estimation with image resolution. However, there is a cumulative error. Therefore, in order to improve the accuracy of self-position estimation over long distances, it is necessary to add a function to eliminate the accumulated error at a certain timing. In this study, we investigated a method of acquiring the robot's absolute coordinates using a camera and AR markers in order to reduce the cumulative error.