An adaptive localization system for first responders

In this contribution a technique for first responders localization support in emergency indoor scenarios is presented. Although localization and mapping has been largely investigated in the field of mobile robotics, it is still a big challenge in emergency response due to the demanding operating conditions. The peculiarity of the proposed system relies on the integration between human operators and robots in a hybrid team. Localization technique takes advantages both from a pre-installed set of landmarks and a self-deployable sensor network composed by tags dropped by rescuers. The main contribution of the paper is the development and the testing of a localization procedure for hybrid rescue teams, able to handle both a priori knowledge, i.e., maps of the environment, with large uncertainties and information updated during rescue tasks.