Abstract
Advancement in the Information and Communications Technology (ICT) has transformed urban environments into modern smart cities by connecting physical devices or sensors to communicate via the Internet of things (IoT) networks. These sensors collect a massive amount of data, which is eventually used in the efficient management of shared assets and resources. In this article, we provide a framework for topological based map building for autonomous robot navigation in smart cities supporting IoT-based technologies. The classical methods of topological based map building do not incorporate the real essence of topology but instead, an illustration is provided where the topology is considered to be a cluster of features. We analyzed the problem of map building from a pure robotic perspective by considering the shape of free-space into a robot navigation path. In the proposed framework, shape-based invariant features are extracted from the free-space, which are used in the construction of topological nodes for robot autonomous navigation. In the context of topological navigation, topology is considered as a structural representation of connecting nodes representing free-space shapes and edges that define their connectivity. To evaluate the performance of the proposed approach, real-time experiments are performed, which show improvements in autonomous navigation in terms of efficiency and maneuverability.
•Advancement in ICT has transformed urban environments into smart cities through IoT.•Classical methods of topological map building do not incorporate the real essence of topology.•Shape-based features are extracted from free-space to form robot navigation nodes.