Indoor Localization Using Computer Vision and Visual-Inertial Odometry

Abstract

Abstract. Indoor wayfinding is a major challenge for people with visual impairments, who are often unable to see visual cues such as informational signs, landmarks and structural features that people with normal vision rely on for wayfinding. We describe a novel indoor localization approach to facilitate wayfinding that uses a smartphone to combine computer vision and a dead reckoning technique known as visual-inertial odometry (VIO). The approach uses sign recognition to estimate the user's location on the map whenever a known sign is recognized, and VIO to track the user's movements when no sign is visible. The advantages of our approach are (a) that it runs on a standard smartphone and requires no new physical infrastructure, just a digital 2D map of the indoor environment that includes the locations of signs in it; and (b) it allows the user to walk freely without having to actively search for signs with the smartphone (which is challenging for people with severe visual impairments). We report a formative study with four blind users demonstrating the feasibility of the approach and suggesting areas for future improvement. State of the Art and Related TechnologyThe key to any wayfinding aid is localization -a means of estimating and tracking a person's location as they travel in an environment. The most widespread localization approach is GPS, which enables a variety of wayfinding tools such as Google Maps, Seeing Eye GPS 1 and BlindSquare, but it is only accurate outdoors. There are a range of indoor localization approaches, including Bluetooth beacons [1], Wi-Fi triangulation 2 , infrared light beacons [2] and RFIDs [3]. However, all of these approaches incur the cost of installing and maintaining physical infrastructure, or of updating the system as the existing infrastructure changes (e.g., whenever Wi-Fi access points change). Dead reckoning approaches such as step counting using inertial navigation [4] can estimate relative movements without any physical infrastructure, but this tracking estimate drifts over time unless it is augmented by absolute location estimates.

Evaluating LevelEd AR: An Indoor Modelling Application for Serious Games Level Design

“…This can cause measurements to become inaccurate as the system may under or over report the distance it has moved since the last marker was placed. This was also noted by [8] and [9] and our work further confirms this. The fourth task was intentionally designed to study the above effect and we believe that the longer the amount of time that is spent on a mapping task then the larger the potential for drift to impact the accuracy, as drift and inaccuracies accumulate over time.…”

“…Similar commercial systems now exist [7], but these focus mainly on the floor plan and not size and placement of objects within the space. There are, however, several academic works utilizing ARKit to develop systems in other domains [8] [9]. Fusco and Coughlan [8] utilized ARKit to develop an indoor localization system for users with visual impairments.…”

“…There are, however, several academic works utilizing ARKit to develop systems in other domains [8] [9]. Fusco and Coughlan [8] utilized ARKit to develop an indoor localization system for users with visual impairments. Whilst Dilek and Erol [9] produced an educational system for generating position-time graphs in real-time.…”

Improved Position Accuracy of Foot-Mounted Inertial Sensor by Discrete Corrections from Vision-Based Fiducial Marker Tracking

“…For ground truth, a similar approach to Fusco et al [ 20 , 35 ] was adopted. The tracked user was followed by a person who recorded a video from an external camera.…”

Fuzzy Logic Type-2 Based Wireless Indoor Localization System for Navigation of Visually Impaired People in Buildings

“…While there is a possibility of using a sort of guide based on an image recognition [4], these systems need to locate some sort of markers or objects and target a camera at them, which is not feasible for a visually impaired person [5]. This problem could be partially solved by machine vision [6]. Most non-wireless systems refer to some sort of proprietary hardware, e.g., microelectromechanical systems (MEMS) [7]; radio-frequency identification (RFID) systems [8] are very rare in artificial laboratory environments and in reality.…”