Loading...
Faculty Sponsor
Dr. Aiyin Jiang, Dr. Zornitza Prodanoff, Dr. Hemani Kaushal
Faculty Sponsor College
College of Computing, Engineering & Construction
Faculty Sponsor Department
Engineering
Location
SOARS Virtual Conference
Presentation Website
https://unfsoars.domains.unf.edu/2021/posters/safe-route-a-mobile-app-based-intelligent-and-personalized-fire-evacuation-system/
Abstract
Project of Merit Winner
All modern buildings have fire evacuation protocols, the most common of which are exit signs to the nearest exit. However, these simple protocols ignore the possibility of unsafe paths and unexpected fire hazards. The goal during fire evacuations is to escape safely and efficiently, but this can be difficult without knowing which routes are safest. A mobile application called Safe Route was created at the University of North Florida (UNF) to provide building occupants with a personalized fire evacuation route helping them to efficiently navigate to the safest exit. Bluetooth Low Energy (BLE) beacons were utilized for the indoor positioning system (IPS) and the network topology that includes a Long Range Wide Area Network (LoRaWAN) infrastructure. A graph network of routes and nodes was designed based upon the provided floor plan of Building 4 at UNF and linked with a routing algorithm to determine the safest route. A safety score was calculated based upon a variety of parameters including temperature, fire growth rate, and carbon dioxide concentration, among others. Dijkstra’s algorithm was then implemented to determine the evacuation route with the best total safety score. Recent tests of Safe Route suggest that evacuation route optimization is effective, and the navigation assistance is accurate. The IPS is still in development to ensure its robustness, and future researchers could continue to improve the associated IPS algorithm. Researchers could also consider commercializing Safe Route or making it open-source for use in fire evacuations anywhere in the world.
Included in
Safe Route: A Mobile App-Based Intelligent and Personalized Fire Evacuation System
SOARS Virtual Conference
Project of Merit Winner
All modern buildings have fire evacuation protocols, the most common of which are exit signs to the nearest exit. However, these simple protocols ignore the possibility of unsafe paths and unexpected fire hazards. The goal during fire evacuations is to escape safely and efficiently, but this can be difficult without knowing which routes are safest. A mobile application called Safe Route was created at the University of North Florida (UNF) to provide building occupants with a personalized fire evacuation route helping them to efficiently navigate to the safest exit. Bluetooth Low Energy (BLE) beacons were utilized for the indoor positioning system (IPS) and the network topology that includes a Long Range Wide Area Network (LoRaWAN) infrastructure. A graph network of routes and nodes was designed based upon the provided floor plan of Building 4 at UNF and linked with a routing algorithm to determine the safest route. A safety score was calculated based upon a variety of parameters including temperature, fire growth rate, and carbon dioxide concentration, among others. Dijkstra’s algorithm was then implemented to determine the evacuation route with the best total safety score. Recent tests of Safe Route suggest that evacuation route optimization is effective, and the navigation assistance is accurate. The IPS is still in development to ensure its robustness, and future researchers could continue to improve the associated IPS algorithm. Researchers could also consider commercializing Safe Route or making it open-source for use in fire evacuations anywhere in the world.
https://digitalcommons.unf.edu/soars/2021/spring_2021/107
Comments
Audio Presentation Transcript:
“Hi, my name is Christian Hayes, and I am an undergraduate research student working alongside Dr. Jiang, Dr. Prodanoff, and Dr. Kaushal. My research project is entitled ‘Safe Route’: A Mobile App-Based Intelligent and Personalized Fire Evacuation System.”
“All modern buildings have fire evacuation protocols, usually to follow exit signs, but these simple evacuation methods are oftentimes insufficient. Escaping a fire requires quick thinking and immediate action, but the safest and quickest way out is not always apparent, and your instincts may lead you down a hazardous path. Safe Route is a mobile application for intelligent and personalized fire evacuation created to help solve this problem by providing real-time navigation assistance for building occupants to escape a fire safely and efficiently. During this study, Safe Route is implemented at the University of North Florida (UNF) to provide a fire evacuation system for the first floor of Building 4.”
“The first objective was to develop an indoor positioning system. Knowing the indoor location of building occupants is essential for personalized fire evacuation. The first step towards developing an indoor positioning system (IPS) was to deploy Bluetooth Low Energy (BLE) beacons to the first floor of Building 4. The beacon used for this study is the Moko H2 iBeacon. To reduce multipath effects from the indoor environment, several sets of experiments were conducted to calibrate the environment and to determine optimal beacon placement before installation. The experimentation includes the observation of received signal strength indicator (RSSI) values at multiple points on the first floor of Building 4. Experimentation was performed to determine the relationship between distance and RSSI, which is shown in Figure 1. A mobile application was created to transmit the RSSI values between each beacon and the building occupant’s smartphone. To estimate the building occupants’ indoor position, a concept called trilateration is used, illustrated in Figure 2. As shown, the occupants must be in range of at least three beacons for trilateration to work properly. When the number of beacons in range of an occupant exceeds three, the indoor positioning accuracy increases because there is more RSSI data to perform calculations on.”
“The second objective was to develop a fire evacuation routing algorithm. The primary purpose of Safe Route’s user interface is to display the safest fire evacuation route. The foundation of determining the safest evacuation route lies upon graph theory and Dijkstra’s algorithm. Dijkstra’s algorithm is used for finding the shortest path between two nodes in a graph network. Before this algorithm can be applied, the indoor floor plan was converted into a network of nodes and routes, as seen in Figure 3. During a fire evacuation scenario, some paths may be more hazardous than others. Therefore, the algorithm had to be customized by giving each path a safety score based upon a variety of factors, including temperature, fire growth rate, and carbon dioxide concentration, among others. Table 1 shows the factors used in the routing algorithm to determine path safety scores. Once a personalized evacuation route is determined for each building occupant, this route will be displayed to them via the Safe Route mobile iOS application, with navigation assistance helping them safely escape.”
“Dijkstra’s algorithm was used to calculate the shortest and safest path from a starting position to one of the exits. The algorithm was modified to calculate the overall safest path based upon a routing table, which is stored in a database. The database and all project files are stored on a UNF webserver. Figure 4 shows the database along with an example of the routing table. A mobile iOS application called Safe Route was created that incorporates both the IPS and fire evacuation routing algorithm. The app was tested on an iPhone XS, as seen by Figure 5. The routing algorithm works well, and the robustness of the IPS algorithm is still being improved. To enhance the user experience, navigation assistance was added by displaying directional arrows, text, and spoken directions.”
“The Safe Route mobile app works in conjunction with carefully positioned BLE beacons to enable an IPS. A custom routing algorithm was built to help guide building occupants to the safest exit in a fire evacuation scenario. The evacuation routing algorithm and navigation assistance were tested by using a custom simulation of predefined user positions to eliminate potential bias from the IPS. The routing algorithm worked smoothly, and accurately guided the user towards the nearest exit. Creating the IPS algorithm is a difficult process partially due to environmental disturbances that affect beacon RSSI values and because of the complex algorithms required. Consequently, a more robust IPS algorithm is still being formulated to enhance the accuracy of user positions, which are estimated based upon RSSI values from nearby beacons. The IPS can be improved by utilizing a more robust algorithm and applying various statistical considerations.”
“This study utilized nine beacons for the IPS, and because each beacon has a limited range, the accuracy of the IPS is less than ideal. Therefore, future researchers should install significantly more beacons as well as continue to improve the IPS algorithm. Once the positioning accuracy improves to a more suitable level, the fire evacuation system could be expanded to span multiple floors or even the entire building. Additionally, the Safe Route app could be commercialized or made open-source for use in fire evacuations anywhere in the world.”
“We would like to acknowledge the support provided by CCEC’s Dean Office, Auld & White Award for carrying out research for this project.”
“This concludes my undergraduate research presentation. Thanks for listening!”