A Hybrid Indoor Positioning Model for Critical Situations Based on Localization Technologies

Abstract

The domains of positioning and tracking have undergone substantial evolution and advancements recently, especially within the concept of the Internet of Things (IoT) and in health care. Unfortunately, neither the current satellite positioning systems nor the standalone cellular systems remain useful for successfully localizing and tracking inside buildings. This paper proposes a new model that could improve the accuracy of localization in indoor environments. In addition, a broad review is conducted to discover the state-of-the-art indoor localization technologies appropriate for disasters and rescue situations. After a comprehensive study, three important technologies that need to be deeply reviewed are identified, which are wireless local area network (WLAN), dead reckoning (DR), and hybrid approaches. Based on these, a novel architecture is introduced that is more convenient to meet the operation of rescuing injured or older people in critical situations, where other technologies might be unavailable or require some extra infrastructures. The proposed model has two modes and selects one of these modes automatically. The first mode assumes the existence of both WLAN signals and smartphone sensors to be used for identifying the position of the object; otherwise, only smartphone sensors will be employed to achieve positioning. Significantly, the designated components and the flow control depicted provide a proper and suitable horizon for the next researchers who desire to develop a new indoor positioning system in this discipline with a low positioning root-mean-squared error on the centimeter scale that can later be incorporated in numerous applications relating to the IoT, health care, and evacuation plans.

Comparative Analysis of Integrated Filtering Methods Using UWB Localization in Indoor Environment

“…We selected a high-contrast area of the robot for tracking. An appropriate tracking algorithm, such as the discriminative correlation filter with a channel and spatial reliability (CSRT) algorithm, was implemented to track the robot within the 512 × 512 images [ 24 ]. The tracking data provide the position within the 512 × 512 image and are converted to the corresponding position in the real environment, which may have different dimensions.…”

A Review of Recurrent Neural Network Based Camera Localization for Indoor Environments

“…To achieve this, we added dropout layers following the final output layer and sub-net to obtain the stochastic pose samples. Because it may predict position errors using uncertainty, the evaluation revealed a significant correlation between uncertainty estimation and location error to increase the re-localization precision of the direct PoseNet for indoor [126] and outdoor environments [127].…”

Comparative Analysis of Integrated Filtering Method Using UWB Localization in Indoor Environments

“…We selected a high-contrast area of the robot for tracking. An appropriate tracking algorithm such as the discriminative correlation filter with a channel and spatial reliability (CSRT algorithm, was implemented to track the robot within the 512 × 512 images [24]. The track ing data provide the position within the 512 × 512 image and are converted to the corre sponding position in the real environment, which may have different dimensions.…”