Proposal of Navigation by Scenarios Using Topological Map
Hiroki SHIMADA, Ryuichi Ueda, Yasuo HAYASHIBARA · The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec) · 2020
We discuss a data format of topological maps and a generation method of the route guidance (scenario). Recent most autonomous mobile robots use a geometrically detailed map for the navigation. On the other hand, humans can move to a destination without such a detailed map. When a human teaches a way to a destination, he/she probably uses the information of landmarks and their connections. We focused on the information that human uses, we discuss the data format of the topological map. Furthermore. We propose a scenario generation method based on the experimental results. We obtained the text of route guidance by a questionnaire and analyze their structures and the most frequent words. And, based on the results, we propose the data format and generation method of the scenario. For verifying the effectiveness of the proposed method, we checked whether the subjects reach the goal based on the generated scenario.