Abstract
This work presents an Indoor Positioning System to estimate the location of people navigating in complex indoor environments. The developed technique combines WiFi Positioning Systems and depth maps, delivering promising results in complex inhabited environments, consisting of various connected rooms, where people are freely moving. This is a non-intrusive system in which personal information about subjects is not needed and, although RGB-D cameras are installed in the sensing area, users are only required to carry their smart-phones. In this article, the methods developed to combine the above-mentioned technologies and the experiments performed to test the system are detailed. The obtained results show a significant improvement in terms of accuracy and performance with respect to previous WiFi-based solutions as well as an extension in the range of operation.
Highlights
Indoor Positioning Systems (IPSs) are techniques employed to calculate the position of people or objects inside buildings
IPSs lean on the calculation of the position of a user in the environments. This localization process can be arranged in four different groups, as stated by Fallah et al in [8]: (1) Dead-reckoning, where position of users is obtained based on a previously estimated or known position through the use of sensors such as accelerometers, magnetometers, compasses, and gyroscopes or using a user’s walking pattern; (2) direct sensing, which determines the location of users through the sensing of identifiers or tags installed in the environment, such as infrared (IR), ultrasound (USID), Bluetooth beacons or bar-codes; (3) the third group uses Triangulation by means of infrared, ultrasound or Radio Frequency Identification (RFID); and (4), Pattern recognition uses data from one or more sensors carried by the user and compares the obtained data with prior collected data
This work presents a method for indoor positioning in complex environments
Summary
Indoor Positioning Systems (IPSs) are techniques employed to calculate the position of people or objects inside buildings. Due to the extended use of WiFi devices (i.e., cellphones) as well as infrastructure (i.e., routers), WiFi-based positioning systems (WPSs) are a suitable opportunity to track the movements of people in indoor environments. These techniques do not deliver very precise results as stated in [1]. A possible solution, based on computer vision, is the use of RGB-D cameras This passive technique has been rapidly developed in the last few years, delivering promising outcomes for people identification [2], positioning [3,4,5] and gesture recognition [6]. Sensors 2017, 17, 2391 on the advantages and limitations of the presented system and suggests future developments based on this method
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
