Camera-based navigation for service robots using pictographs on the crossing point

Abstract

In our current research, we have developed a novel navigation system for indoor service robots using pictographs. Pictographs are widely used in a general environment with human such as offices, stations and welfare care facilities. Using them as a landmark, our system does not require special markers in the navigation area of the robot. Our robot is only equipped with a low-cost camera and uses SURF (Speeded Up Robust Feature) algorithm for detecting pictographs. In this paper, we propose a novel robot navigation scheme using the result of pictograph detection. Our two key ideas are: One is the robot's position and its advance direction estimation scheme using the pictograph detection. The other topic is the navigation scheme using the pictograph equipped on the crossing point. Usually, the pictograph shows the direction to the destination such as an exit, a toilet and an information center. Therefore, the pictograph tends to be equipped in a fork in the load and our system uses its characteristic for the robot's navigation. Our system integrates the position estimation results by the pictograph detection and the results by odometry, and derives the path to the goal position of the robot. We verify the effectiveness of our scheme by experiment with our prototype.