Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.

  

Real-time path planning for emergency building evacuation

Student(en):

Betreuer:

Summers Sean
Beschreibung:

Most inhabited buildings are equipped with static escape routes, illustrated by maps sparsely posted on the walls and usually indicating a route to the nearest exit. In an emergency (for example, a fire emergency), it is assumed that the building inhabitants will follow the pre-specified routes. However, this approach to building evacuation is both naive and outdated. Often, building inhabitants do not notice the escape routes and in an emergency do not take the time to study them. Further, the location of building vulnerability and its evolution in time may put many pre-determined escape routes at risk. Thus, in an era of smart buildings, it is only natural that the approach to building evacuation be updated as well. It is therefore the goal of this project to develop a framework for building evacuation that uses information obtained and updated in real time (e.g. location of building vulnerabilities) to update the escape routes and broadcast these updates to the inhabitants (e.g. arrows on building walls).

In this project, the student will use recent advances in stochastic reachability to develop a control strategy for emergency building evacuation. Specifically, the student will consider the emergency to be a fire that is spreading throughout the building, hence the building must be evacuated. Based on findings in the literature, the student will develop a stochastic model for fire propogation throughout the building. Then, using the stochastic model of fire propogation and the theory of stochastic reachability with random sets, the student will synthesize a control strategy (real-time path planning) that maximizes the probability that the inhabitants of the building are evacuated safely.

Weitere Informationen
Professor:

John Lygeros
Projektcharakteristik:

Typ:
Art der Arbeit:
Voraussetzungen: Signals and Systems I and II, familiar with MATLAB programming.
Anzahl StudentInnen:
Status: open
Projektstart:
Semester: