CAJ | 학술논문

The human following becomes one of the significant procedure in human-friendly navigation of mobile robots.Many potential applications of human-following technology are developed lately.This paper suggests a method for a mobile robot to detect human legs and to follow the human by using a single laser range finder (LRF).We extract four simple attributes of human legs.We employ a support vector data description (SVDD) scheme in order to define the boundary of extracted attributes mathematically.To deal with occlusion,we designed an efficient human walking model to have robust tracking.The proposed approaches were successfully confirmed by carrying out various experiments.