Today, a lot of manual work is required to create an indoor navigation solution, both for creating indoor positioning and for creating indoor navigation with maps and routes. In this project, we want to develop algorithms and an indoor navigation prototype where self-learning positioning and machine learning automatically identifies routes indoors for indoor navigation.
Global Indoor Navigation (GIN) is an industrial research project that aims to develop a new self-learning indoor navigation that automatically characterizes buildings through the use of the service on regular mobile phones. The characterization is intended to indicate: number of floors, where stairs / elevators / escalators are located, create a simple indoor map, identify the type of building (housing, office, shop) and type of rooms. With this characterization, it is possible to create a number of valuable services eg. indoor navigation, security solutions, locating people in need, finding assets, detecting building changes, identifying bottlenecks, improving accessibility and other efficiency solutions.
The project is based on technology from the previous research project “Global Indoor Positioning in 3D” where, using new SLAM technology, it is possible to automatically create a good indoor positioning based on existing WiFi access points and Bluetooth beacons. This SLAM technology together with a new WiFi standard, 802.11mc, with meter precision in the indoor positioning, offers completely new opportunities for new innovation in indoor navigation for the future.
Our vision is to create a self-learning global indoor navigation that works in all buildings around the world.