Vivid: Augmenting Vision-Based Indoor Navigation System With Edge Computing

Abstract

Indoor localization and navigation have a great potential of application, especially in large indoor spaces where people tend to get lost. The indoor localization problem is the fundamental of an indoor navigation system. Existing research and commercial efforts have leveraged wireless-based approaches to locate users in indoor environments. However, the predominant wireless-based approaches, such as WiFi and Bluetooth, are still not satisfactory, either not supporting commodity devices, or being vulnerable to environmental changes. These issues make them hard to deploy and maintain. In this paper, we present Vivid, a mobile device-friendly indoor localization and navigation system that leverages visual cues as the cornerstone of localization. By leveraging the computation power at the extreme internet edges, Vivid to a large extent overcomes the difficulties brought by resource-intensive image processing tasks. We propose a grid-based algorithm that transforms the feature map into a grid, with which finding the path between two positions can be easily obtained. We also leverage deep learning techniques to assist in automatic map maintenance to adapt to the visual changes and make the system more robust. With edge computing, user privacy is preserved since the visual data is mainly processed locally and detected dynamic objects are removed immediately without saving to databases. The evaluation results show that: i) our system easily outperforms the existing solutions on COTS devices in localization accuracy, yielding decimeter-level error; ii) our choice of the system architecture is scalable and optimal among the available ones; iii) the automatic map maintenance mechanism effectively ameliorates the localization robustness of the system.INDEX TERMS Last mile delivery, loTs-based indoor localization and navigation, edge computing for loTs sensors.

Cloud Platforms for Context-Adaptive Positioning and Localisation in GNSS-Denied Scenarios—A Systematic Review

“…Current indoor navigation platforms use this technology along with AR techniques in order to provide information of the environment [ 59 , 114 ]. Given the significant demand for computing resources, most of the processes are moved from mobile devices to the Cloud [ 115 ] as we saw in previous paragraphs.…”

“…Unlike technologies like Wi-Fi and BLE, with computer vision it is possible to achieve decimetre-level accuracy, for instance, in [ 115 ] the authors acquired a localisation error of 60 (approx.) in an area of 7.2 m × 5.2 m. Additionally, the researchers tested their system in three different scenarios of different sizes and obtained similar results.…”

“…Efficient Computation [ 45 , 48 , 52 , 61 , 68 , 74 , 85 , 94 , 102 , 115 ]: It consists of improving the methods or algorithms used in mobile devices and the Cloud in order to decrease the use of computational resources. To reduce the computational load in mobile devices, the authors offload specific processes to the Cloud or other computational paradigms (see Section 4.1 ).…”

Ground-Level Mapping and Navigating for Agriculture Based on IoT and Computer Vision

“…Details are described in Algorithm 1 with an example in Fig. 3-b using monocular ORBSLAM2 [35]. A redundant intermediate output is a normal vector of the mapped plane, and it could be avoided by applying a mesh map that labels the accessibility of areas.…”

“…Hence, it is important to use energy efficient hardware and develop energy and computational efficient software. As discussed in our previous work [35], current monocular-based SLAM systems are not suitable to run on portable devices for our purpose. In terms of computing hardware of this IoT system, different hardware configurations were chosen for Cloud server and Edge nodes as shown in Table III.…”