Optical Camera Communication system for three-dimensional indoor localization

Abstract

Nowadays, indoor localization methods are intensively researched due to their potential use in location-based services, robots' navigation and applications for Industrial Internet of Things (IIoT). The tendency of including small devices with Light Emitting Diodes (LED) for decoration purposes and the emergence of better low-cost cameras in the market, impulse the use of Optical Camera Communication (OCC) for Indoor Positioning Systems (IPS). In this work a real-time two-step 3D Indoor Positioning System for multiple devices using OCC and LED beacons is proposed. The beacons are positioned at a predefined static distance from the camera, forming a virtual reference plane that represents the frame of the real world elements' positions captured in a photo. A simplified version of the trigonometric equations for transforming 2D pixel projection into real distances is applied to obtain the vertical and horizontal separation between the objects and the beacons. While the relation of the beacons' and the objects' projection size is used to determine the distance of each object to the camera. These distances together allowed to calculate the 3D location of the elements by trilateration. Experimental validation of the method was performed, and the results showed a mean location error of 1.24 cm for the three dimensions with an average process time of 18.2 ms per frame for four simultaneous targets correctly identified, at a distance of 2.00 m.