Development of a Person-Searching Algorithm Using an Omnidirectional Camera and LiDAR for the Tsukuba Challenge

摘要

In this paper, we describe the development of a person-searching algorithm that used a combination of an omnidirectional camera and LiDAR (light detection and ranging), for the Tsukuba Challenge. We used an omnidirectional camera because they can observe wide angles compared with conventional cameras. To reduce computational cost for image processing, we focused on color information only. An omnidirectional camera is used to detect the direction of specifically colored personal wear. Once the mobile robot finds the direction of the specific color, the LiDAR is used to detect an auxiliary signboard based on the known width around the detected color direction to confirm the target. The validity of the proposed method was confirmed by preliminary experiments that emulated Tsukuba Challenge environments.