Skip to main content
eScholarship
Open Access Publications from the University of California

Real-time WiFi localization of heterogeneous robot teams using an online random forest

  • Author(s): Benjamin, B
  • Erinc, G
  • Carpin, S
  • et al.
Abstract

© 2015, Springer Science+Business Media New York. In this paper we present a WiFi-based solution to the localization and mapping problem for teams of heterogeneous robots operating in unknown environments. By exploiting wireless signal strengths broadcast from access points, a robot with a large sensor payload creates a WiFi signal map that can then be shared and utilized for localization by sensor-deprived robots. In our approach, WiFi localization is cast as a classification problem. An online clustering algorithm processes incoming WiFi signals that are then incorporated into an online random forest (ORF). The algorithm’s robustness is increased by a Monte Carlo localization algorithm whose sensor model exploits the results of the ORF classification. The proposed algorithm is shown to run in real-time, allowing the robots to operate in completely unknown environments, where a priori information such as a blue-print or the access points’ location is unavailable. A comprehensive set of experiments not only compares our approach with other algorithms, but also validates the results across different scenarios covering both indoor and outdoor environments.

Many UC-authored scholarly publications are freely available on this site because of the UC Academic Senate's Open Access Policy. Let us know how this access is important for you.

Main Content
Current View