Ceiling vision-based active SLAM framework for dynamic and wide-open environments

Abstract

A typical indoor environment can be divided into three categories; office (or room), hallway, and wide-open space such as lobby and hall. There have been numerous approaches for solving simultaneous localization and mapping (SLAM) problem in office (or room) and hallway. However, direct application of the existing approaches to wide-open space may be failed, because it has some distinguished features compared to other indoor places. To solve this problem, this paper proposes a new ceiling vision-based active SLAM framework, with an emphasis on practical deployment of service robot for commercial use in dynamically changing and wide-open environments by adopting the ceiling vision. First, for defining ceiling feature which can be extracted regardless of complexity of ceiling pattern we introduce a model-free landmark, i.e., visual node descriptor, which consists of edge points and their orientations in image space. Second, a recursive `explore and exploit' is proposed for autonomous mapping. It is recursively performed by spreading out mapped area gradually while the robot is actively localized in the map. It can improve map accuracy due to frequent small loop closing. Third, a dynamic edge link (DEL) is proposed to cope with environmental changes in the map. Owing to DEL, we do not need to filter out corrupted sensor data and to distinguish moving object from static one. Also, a self-repairing map mechanism is introduced to deal with unexpected installation or removal of inner structures. We therefore achieve long-term navigation. Several simulations and real experiments in various places show that the proposed active SLAM framework could build a topologically consistent map, and demonstrated that it can be applied well to real environments such as wide-open space in a city hall and railway station.