• Overview
• Visitor Info
• People

• Publications

• Courses
• GA/SA/MA

• News
• Seminars

Indoor Localisation for Building Evacuation

Firefighting is a challenging dangerous task, involving high risks combined with fast decision-making. In most cases firefighters use their experience combined with a limited knowledge of the burning structure and geometry to decide the path to take for rescuing individuals and containing the fire. In this thesis, we develop the backbone structure for a navigation tool to help firefighters make optimized decisions using real-time data and optimal control algorithms.

In order to communicate optimized navigation strategies to firefighters in real-time, we need to build a system that incorporates indoor localization and communicates position measurements at fast sampling times to a central computer. The central computer then communicates navigation strategies based on this location information to hand-held devices, such as cell phones. We have previously built an award-winning indoor localization technology at our institute. Our goal in this project is to close the loop of measurement and localization with the communication of navigation strategies to users.

The eventual goal of this project is to integrate advanced optimal control algorithms in determining safe and optimized navigation strategies for firefighters and other first responders. Therefore, this project plants the initial seeds for creating an experimental platform for verification of the control theory developed for this crucial application.

Completion of this project involves the following tasks:

• Setup of a robust indoor localization platform, building on our in-house existing UWB-based localization system,
• Communicate the location information to a central computer,
• Communicate navigation strategies, determined based on location input, to a mobile phone,
• Develop an app that displays the navigation strategies on the mobile phone.

You will have the chance to work on a cutting-edge indoor localisation technology. You will gain experience in setting up the hardware for a real-time control experiment. You will implement a Kalman filter base on real application data.