Junseo Kim, Jill Aghyourli Zalat, Yeganeh Bahoo, and Sajad Saeedi

Toronto Metropolitan University (TMU)

With the rising prominence of WiFi in common spaces, efforts have been made in the robotics community to take advantage of this fact by incorporating WiFi signal measurements in indoor SLAM (Simultaneous Localization and Mapping) systems, termed WiFiSLAM. This paper describes recent work in the development of WiFiSLAM systems and addresses the challenges currently faced in achieving WiFi-based mapping used in typical SLAM applications. A discussion of the usage of alternative WiFi-based systems is conducted, with a focus on the field of research into k-visibility.  We introduce the novel concept of an inverse k-visibility algorithm and its application in using a free-space prediction model to achieve geometric mapping of an environment. A suggested approach to build upon this model is given, with simulated and experimental results implemented to demonstrate the proposed inverse k-visibility-based approach.

In this paper, we use concepts from the field of k-visibility to devise a novel approach for achieving 2D geometric mapping of an indoor space using only WiFi signal-strength measurements and trajectory information. We propose the inverse k-visibility algorithm, which uses probabilistic modeling of known k-visibility information to estimate an explicit model of the environment, which we call a Structure from WiFi (SfW) map. 

The main contribution of the work is bringing k-visibility concepts into robotics mapping problems and proposing a mapping algorithm that maps most of the free space using WiFi RSSI signals only without relying on sensors such as lidar, radar, or camera. Mapping free space is significant, as it allows the robots to plan paths without colliding with obstacles. Evaluation of the work in simulation and real-world demonstrates the significance of the method. 

@inproceedings{junseokimwifi24,

author={Junseo Kim and Jill Aghyourli Zalat and Yeganeh Bahoo and Sajad Saeedi},

booktitle={American Control Conference (ACC)},

title={Structure from WiFi (SfW): RSSI-based Geometric Mapping of Indoor Environments},

year={2024},

note={Accepted.},

}