An Accurate Visual-Inertial Integrated Geo-Tagging Method for Crowdsourcing-Based Indoor Localization

Abstract

One of the unavoidable bottlenecks in the public application of passive signal (e.g., received signal strength, magnetic) fingerprinting-based indoor localization technologies is the extensive human effort that is required to construct and update database for indoor positioning. In this paper, we propose an accurate visual-inertial integrated geo-tagging method that can be used to collect fingerprints and construct the radio map by exploiting the crowdsourced trajectory of smartphone users. By integrating multisource information from the smartphone sensors (e.g., camera, accelerometer, and gyroscope), this system can accurately reconstruct the geometry of trajectories. An algorithm is proposed to estimate the spatial location of trajectories in the reference coordinate system and construct the radio map and geo-tagged image database for indoor positioning. With the help of several initial reference points, this algorithm can be implemented in an unknown indoor environment without any prior knowledge of the floorplan or the initial location of crowdsourced trajectories. The experimental results show that the average calibration error of the fingerprints is 0.67 m. A weighted k-nearest neighbor method (without any optimization) and the image matching method are used to evaluate the performance of constructed multisource database. The average localization error of received signal strength (RSS) based indoor positioning and image based positioning are 3.2 m and 1.2 m, respectively, showing that the quality of the constructed indoor radio map is at the same level as those that were constructed by site surveying. Compared with the traditional site survey based positioning cost, this system can greatly reduce the human labor cost, with the least external information.

Automatic radio map creation in a fingerprinting‐based BLE/UWB localisation system

“…Compared to the solutions presented in the literature, the proposed method performs well as it allows to achieve localisation of comparable or better accuracy. The obtained results were more accurate than in case of WiFi‐based systems using Gaussian processes regression, where the average errors were about 2.3–2.4 m [14], 3 m [11] and 5 m [15]. A system of similar accuracy was presented in [19], where separate radio map for peak hours, when the covered area is typically crowded, was used.…”

“…In passive crowdsourcing, the basic principle is the same but users' locations are determined automatically by using other sensors or techniques. Example of the crowdsourcing solution, in which the users location is determined using smartphone's inertial sensors can be found in [10, 11]. There are also some works, which localise the users using other accessible localisation systems.…”

Continuous Indoor Visual Localization Using a Spatial Model and Constraint

“…In [46], iMoon built a 3D model of an indoor environment by using SFM technology that supports image-based localization. Reference [23] employed an SFM method to estimate the trajectory of a moving camera in an indoor environment. However, a problem is that the initial location of a camera should be given as an input for SFM-based trajectory recovery.…”

“…As a well-known imaging technology, the SFM method can be used to recover the relative camera pose and 3D structure from a set of camera images. Previous studies have utilized the SFM method to recover the geometry of trajectories in indoor spaces [22], [23]. An advantage of the SFM method is that the heading estimation error of the SFM is significantly smaller than that of the PDR-based estimation (from the gyroscope).…”

APS: A Large-Scale Multi-modal Indoor Camera Positioning System

“…Some conventional methods in this regard, either use RGB images along with a depth-assisted camera [ 3 , 10 , 18 , 40 , 42 , 43 ], or they employ SIFT-based algorithms [ 19 , 32 , 38 ]. In many realistic scenarios however, the depth-based camera or bluetooth or WiFi signals are not available [ 4 , 22 , 33 ], at all.…”