Human comfort following behavior for service robots

Abstract

Human following is one of the most important tasks for a service robot sharing its workspace with humans. The human following behavior design for service robots is considered in this paper, with applications to human-robot interaction and human companion in indoors environments. In this work, we present the necessary of future motion expectation and natural interaction for human following model. The key advances in this paper are: an intelligent following system based on user comfortable feeling and put forward the comprehensive performance evaluation index of following system. As a result, the evaluation shows that the proposed model considering human comfort performs better than a traditional model that only maintain a constant distance with the human.