Development of Blocked Route Signboard Detection Algorithm for Autonomous Mobile Robot Combining Omnidirectional Camera and 3D-LiDAR

This paper describes the development of an algorithm for detecting blocked route signboard in an autonomous mobile robot using a combination of omnidirectional camera and 3D-LiDAR. A signboard recognition technique is used to detect a blocked route signboard. An omnidirectional camera and 3D-LiDAR are used for stable signboard detection. The omnidirectional camera is mainly used for stable color detection regardless of outdoor sunlight position, while the 3D-LiDAR is used for the retroreflective painted areas used in the signboard and for detecting the shape of the known signboard dimension by point cloud extraction. The validity of the proposed algorithm was confirmed by detecting blocked route signboard with the same dimensions as those defined in the Tsukuba Challenge rule.