Indoor Mobile Robot Socially Concomitant Navigation System

Abstract

With the progress of science and technology and the development of economy, more and more robots have entered the family and become good helpers and partners in people’s daily life. As a good partner of human beings, the robot is endowed with humanized emotion and decision-making core, and has harmonious and natural social interaction ability. Among them, the socially concomitant navigation function is a very important social function. For indoor mobile robot to navigate socially with its human companions, a Single-Object Tracking deep-sort(SOT deepsort)algorithm is proposed, and a fast, stable and simple socially concomitant navigation system is proposed. In order to make the robot more socialized, questionnaires are designed to investigate the comfortable distance of the robot in the process of following its companion. Endow the robot appropriate behavior when traveling with human companions, so that the robot can play the role of accompanying the pedestrian and complete the task of following the companion. Finally, the proposed system is tested on Fabo mobile robot. The experimental results show that the following accuracy of this method is high and meets the real-time requirements.