画像検索に基づく定位技術を夜間利用するための移動ロボット動作アルゴリズム

―暗所でのUSBカメラを用いた定位―

抄録

As autonomous mobile robots have become increasingly popular in our living spaces in recent years, we considered the need to reduce power consumption when taking into account the long hours of operation of the robots. Therefore, we considered using images obtained by image sensors and the universal map (UMap) for indoor location estimation. However, for robots that are operated for long periods of time, the image sensor may not function at night. Therefore, we verified whether line segment information necessary for position estimation could be obtained by adjusting the exposure time and gain of the sensor. From the experimental results, we confirmed that line segments of buildings can be obtained by adjusting the exposure time and gain of the image sensor. This suggests the possibility of achieving high accuracy while reducing power consumption in position estimation considering long hours of operation of autonomous mobile robots.