# WiFi-SLAM Using Gaussian Process Latent Variable Models

Brian D. Ferris
Dieter Fox
Neil D. Lawrence, University of Sheffield

in Proceedings of the 20th International Joint Conference on Artificial Intelligence (IJCAI 2007), pp 2480-2485

#### Abstract

WiFi localization, the task of determining the physical location of a mobile device from wireless signal strengths, has been shown to be an accurate method of indoor and outdoor localization and a powerful building block for location-aware applications. However, most localization techniques require a training set of signal strength readings labeled against a ground truth location map, which is prohibitive to collect and maintain as maps grow large. In this paper we propose a novel technique for solving the WiFi SLAM problem using the Gaussian Process Latent Variable Model (GP-LVM) to determine the latent-space locations of unlabeled signal strength data. We show how GP-LVM, in combination with an appropriate motion dynamics model, can be used to reconstruct a topological connectivity graph from a signal strength sequence which, in combination with the learned Gaussian Process signal strength model, can be used to perform efficient localization.

  @InProceedings{ferris-wifi07, title = {WiFi-SLAM Using Gaussian Process Latent Variable Models}, author = {Brian D. Ferris and Dieter Fox and Neil D. Lawrence}, booktitle = {Proceedings of the 20th International Joint Conference on Artificial Intelligence (IJCAI 2007)}, pages = {2480}, year = {2007}, editor = {Manuela M. Veloso}, month = {00}, edit = {https://github.com/lawrennd//publications/edit/gh-pages/_posts/2007-01-01-ferris-wifi07.md}, url = {http://inverseprobability.com/publications/ferris-wifi07.html}, abstract = {WiFi localization, the task of determining the physical location of a mobile device from wireless signal strengths, has been shown to be an accurate method of indoor and outdoor localization and a powerful building block for location-aware applications. However, most localization techniques require a training set of signal strength readings labeled against a ground truth location map, which is prohibitive to collect and maintain as maps grow large. In this paper we propose a novel technique for solving the WiFi SLAM problem using the Gaussian Process Latent Variable Model (GP-LVM) to determine the latent-space locations of unlabeled signal strength data. We show how GP-LVM, in combination with an appropriate motion dynamics model, can be used to reconstruct a topological connectivity graph from a signal strength sequence which, in combination with the learned Gaussian Process signal strength model, can be used to perform efficient localization.}, key = {Ferris:wifi07}, linkpdf = {http://www.ijcai.org/papers07/Papers/IJCAI07-399.pdf}, group = {gplvm,dimensional reduction} }
 %T WiFi-SLAM Using Gaussian Process Latent Variable Models %A Brian D. Ferris and Dieter Fox and Neil D. Lawrence %B %C Proceedings of the 20th International Joint Conference on Artificial Intelligence (IJCAI 2007) %D %E Manuela M. Veloso %F ferris-wifi07 %P 2480--2485 %R %U http://inverseprobability.com/publications/ferris-wifi07.html %X WiFi localization, the task of determining the physical location of a mobile device from wireless signal strengths, has been shown to be an accurate method of indoor and outdoor localization and a powerful building block for location-aware applications. However, most localization techniques require a training set of signal strength readings labeled against a ground truth location map, which is prohibitive to collect and maintain as maps grow large. In this paper we propose a novel technique for solving the WiFi SLAM problem using the Gaussian Process Latent Variable Model (GP-LVM) to determine the latent-space locations of unlabeled signal strength data. We show how GP-LVM, in combination with an appropriate motion dynamics model, can be used to reconstruct a topological connectivity graph from a signal strength sequence which, in combination with the learned Gaussian Process signal strength model, can be used to perform efficient localization. 
 TY - CPAPER TI - WiFi-SLAM Using Gaussian Process Latent Variable Models AU - Brian D. Ferris AU - Dieter Fox AU - Neil D. Lawrence BT - Proceedings of the 20th International Joint Conference on Artificial Intelligence (IJCAI 2007) PY - 2007/01/01 DA - 2007/01/01 ED - Manuela M. Veloso ID - ferris-wifi07 SP - 2480 EP - 2485 L1 - http://www.ijcai.org/papers07/Papers/IJCAI07-399.pdf UR - http://inverseprobability.com/publications/ferris-wifi07.html AB - WiFi localization, the task of determining the physical location of a mobile device from wireless signal strengths, has been shown to be an accurate method of indoor and outdoor localization and a powerful building block for location-aware applications. However, most localization techniques require a training set of signal strength readings labeled against a ground truth location map, which is prohibitive to collect and maintain as maps grow large. In this paper we propose a novel technique for solving the WiFi SLAM problem using the Gaussian Process Latent Variable Model (GP-LVM) to determine the latent-space locations of unlabeled signal strength data. We show how GP-LVM, in combination with an appropriate motion dynamics model, can be used to reconstruct a topological connectivity graph from a signal strength sequence which, in combination with the learned Gaussian Process signal strength model, can be used to perform efficient localization. ER - 
 Ferris, B.D., Fox, D. & Lawrence, N.D.. (2007). WiFi-SLAM Using Gaussian Process Latent Variable Models. Proceedings of the 20th International Joint Conference on Artificial Intelligence (IJCAI 2007) :2480-2485