1P1-J04 Navigation of Autonomous Mobile Robot that is Robust against Topological Change of Environment
Rihiro KAMEI, Yuichi Tazaki, Tatsuya Suzuki · The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec) · 2015
In this paper we present a robust navigation method under static environment, where there are unknown obstacles, using a topological map based on Generalized Voronoi Diagrams (GVD). By getting observation information by Laser Range Finder (LRF), the mobile robot can be navigated in the topological map. In case that unknown obstacles are recognized, robust navigation is done by considering its topological effect. We assume that topological changes by unknown obstacles are classified in two patterns. One is that there is an unknown obstacle "in a crossroad". The other is "on a passage". Every time the mobile robot encounters an unknown obstacle, the algorithm classifies a local topological change in one of two patterns. Based on the pattern classified, the algorithm plans a local alternate path. An experiment in the indoor environment by the mobile robot shows the good performance of the algorithm for navigation.