Indoor and Seamless Positioning
and Navigation (J-ISPIN)

Twentythree ultra-wideband anchors collaborate to calibrate themselves in challenging industrial conditions with limited connectivity. The algorithm is able to localize the anchors with an mean absolute error of only 21.6 cm.

Necessity of environment detection in a seamless localization scenario, where the mobile object moves in different environments and uses various positioning methods. For example:
1) crowded urban area, where (s)he uses 5G signals for positioning,
2) indoor building, where (s)he utilizes camera and map-based method for localization,
3) open area, where the pedestrian deploys GNSS receivers to locate itself.

AI improves wearable inertial device-based pedestrian navigation. This paper classifies these AI methods into two categories based on signal segmentation methods: human gait and sampling frequency-driven. To complete the analysis, SELDA (category 1), RONIN (category 2) and a non-AI gait-driven method SmartWalk were evaluated in a 2,17 km long open access dataset, representative of the diversity of pedestrians’ mobility surroundings.