IPIN 2021: Indoor Positioning Using the OpenHPS Framework

Our paper on Indoor Positioning Using the OpenHPS Framework was presented today at the 11th International Conference on Indoor Positioning and Indoor Navigation (Lloret de Mar, Spain). In this paper we introduced OpenHPS as a framework that is capable of handling an indoor positioning use case. We demonstrate this using a flexible and configurable process network that handles indoor positioning using existing techniques and algorithms. The flexibility of our framework allows developers to tweak every node in the process network to optimally sample the data.

Additional resources related to this paper can be found at the bottom of this blog post

positioning model process network consisting of an offline application (for calibration), online application (for users) and a server that does the fingerprinting and multilateration

In the demonstrator that we created we use the @openhps/socket, @openhps/react-native, @openhps/fingerprinting, @openhps/rf, @openhps/imu and @openhps/mongodb modules on top of our core component. Two applications are created, one for the offline stage (i.e. used for calibrating and setting up the system), one for the online stage (i.e. used by users in the production environment) and finally a server to perform the positioning. Our online application uses Wi-Fi, BLE and IMU sensor data to obtain a position.

We recorded a new dataset in our research lab with several training an test data points that can be used to test the positioning model. We also included several trajectories that can use the IMU data of our system to determine more up-to-date position estimates.