第六章 結論與建議
第二節 建議
本研究在結合電子火災逃生地圖與避難引導系統研究方面提出下列具體建議。
建議一
立即可行建議：推動「大型公共開放空間之智慧型緊急避難導引系統」
主辦機關：內政部建築研究所
本研究「結合電子火災逃生地圖與避難引導系統」之研究成果，適合於特定 空間之使用，如一般的辦公大樓，在此特定的空間中人員均須配戴員工證(RFID 標籤)，本計畫之研究成果可執行 RFID 即時人員定位功能，數位電子看板並可依 不同建置位置提供不同的逃生導引資訊，但在大型公共開放空間卻無法精準地計 算出人員之數量與分佈狀況，如何運用影像辨識技術進行避難者標示及計數，並 針對環境中避難者分布情形以及環境狀態即時建構一個可有效且具人員分流能 力之智慧型緊急避難導引雛型系統，可將避難者快速導引往出口避難，並以持續 更新導引資訊方式即時依據環境狀態反應於導引路徑上，「大型公共開放空間之 智慧型緊急避難導引」將可將貴所近兩年來所持續發展之內容成為一套可廣泛應 用於多種不同類型建築物且不受環境型態影響之緊急避難導引系統。
建議二
立即可行建議:辦理智慧型防火避難引導系統相關成果發表講習會 主辦機關:內政部建築研究所
協辦機關:營建署、消防署、台灣建築中心、各相關建築師、消防設備師公會 透過成果發表會方法，將智慧型避難導引系統開發之成果，推廣至建築師或 消防設備師士等團體，提供各界參採應用；亦可將本研究之電子看板結合避難導 引系統之模式，運用於使用單位之避難演練，強化防火避難行動。
75
建議三
中期建議:申辦相關專利 主辦機關:內政部建築研究所 協辦機關:中華大學
將研究之成果或智慧財產，整理具有新穎性、進步性、產業利用性等之可專 利部分，研議專利申請書，向智慧財產局申請專利，共同推動研究成果。
76
77
附錄一 已發表之論文
An Intelligent Active Alert Application on Handheld Devices for Emergency
Evacuation Guidance
Chung-Chuo Wu
1, Kun-Ming Yu
1, Shao-Ting Chine
1, Shao-Tsai Cheng
2, Yuan-Shao Huang
1, Ming-Yuan Lei
3, Jiann-Horng Lin
31Department of Computer Science and Information Engineering, Chung Hua University
2Department of Construction Management, Chung Hua University
3Architecture and Building Research Institute, Ministry of the Interior
1,2Hsinchu, Taiwan
3New Taipei City, Taiwan
m10002045@chu.du.tw, yu@chu.edu.tw, blingromance@gmail.com, shaotsai@chu.edu.tw,
m10002044@chu.edu.tw, alec@abri.gov.tw, jhlin@abri.gov.tw
Abstract— When an emergency occurs, like a fire guidance when emergencies happen, an emergency crowd guidance system was designed and implemented. In the system, a wireless sensor network, RFID, and a smart handheld device were integrated to provide a real-time, active, intelligent guidance system for fire evacuation. Moreover, an emergency guidance algorithm was run in the central control module of the proposed system displaying the evacuation route on mobile devices.
Also, in the proposed system, the user can obtain information of the environment and get evacuation guidance with a well-marked arrow on a mobile device.
Keywords— Wireless sensor networks, RFID location, emergency crowd guidance application
I. Introduction
According to an estimation of the United States Census Bureau, there were 7.057 billion people in the world at the end of 2012. It leads to buildings becoming higher and more complicated in cities. Because of buildings being complicated, people might evacuate in the wrong direction when an emergency, like fire, occurs. So it is imperative for evacuation to be as effective as possible in an emergency. To navigate people out
of danger, accurate information is needed.
First, for emergency guiding purposes the evacuation path in a building needs to be determined. There is adequate research regarding emergency evacuation, some of it uses guidance algorithms; others focus on guidance systems, Fire Escape Equipment location, etc. In 2006, Tseng et al [13] proposed an emergency guidance algorithm using a wireless sensor network to prove the algorithm. Cheng et al [3] propose a load-balancing emergency guiding system (LEGS), which uses corridor capacity and time to balance the evacuation time of each path.
Second, various parameters of environment are needed for evacuation guidance algorithms to operate the path. The most popular way of collecting information about environment is using wireless sensor networks (WSNs). WSN can be used for monitoring physical environment conditions such as luminance, humidity, temperature, etc. The application includes air pollution monitoring, building monitoring, etc. It is a mature and widely used technology.
Third, to navigate people out of a building the location of each user needs to be known. The algorithm also needs information regarding the number of people in the building. RFID indoor location is a popular and precision technique used for this. It uses an RFID Reader and RFID Tag to contrast the quality of signals to exactly locate each person.
78 In this study, a system with location, environment information, algorithm, database, and application was designed to establish a real-time emergency crowd guidance system.
The system was set up in a hotel to test.
This paper is organized as follows: In section 2, related work on emergency evacuation, wireless sensor networks, and indoor locations are given. The system design is briefly discussed in section 3. Section 4 discusses the design of the application system. Section 5 gives the system Prototype Implementation.
II. Related Worls
In this system design, indoor location, environment sensing, emergency evacuation guidance technology, and handheld application for emergency evacuation guidance algorithm were used to navigate people out of danger.
For determining location, the global positioning system (GPS) is usually used outdoors; however, it cannot be used indoors because the signal is too weak. Getting precision in indoor location sensing is a problem to be solved. For the past few years, many studies using different methods [2, 6, 12] have been done.
Different types of hardware were used to find indoor location like Wi-Fi, ZigBee, Bluetooth, and RFID. Because of its superior precision, RFID has become popular for determining indoor location. L. M. Ni. et al [10] propose an indoor RFID location method, LANDMARC, it uses many RFID Readers and Tags for location. To reduce RFID hardware and increase precision M.
G. Lee et al [9] proposed a VSLS system in 2010. precision.
Wireless Sensor Networks (WSNs) is a popular subject in environmental sensing and electrical equipment control. WSNs are usually applied with ZigBee communication protocol for its low data rate, long battery life, and supporting multiple networks as star, tree, and meth. ZigBee wireless sensor network is applied in many research and ecology environment monitoring projects [5, 7, 14]. In 2009, Choi et al [1] used wireless sensor network to monitor and control indoor air quality in a subway station. And in 2010, Yu et al [16] also proposed a context-aware living environment (CALE) using the ZigBee sensor network to collect the information and control the electrical equipment in an office.
The emergency evacuation research [8, 11, 15]
has many different aspects including macro-phase, micro-phase, crowd guidance, single guidance, etc. There is some research on the shortest and safest evacuation paths. In 2012, Zhou et al [17] proposed the multiple streaming crowd guidance (MSCG). The MSCG blends the macro- and micro-phases and gives each person individual evacuation guidance. In MSCG, the evacuation path operates according to temperature, humidity, light, CO, CO2, area capacity, fire propagation, etc. The concept of water flow, from high to low, is used in guiding people. According to the MSCG algorithm, people are navigated separately according to the distance and capacity of exits.
III. System Design
In this study, an emergency crowd guidance system was designed to navigate people out of dangerous buildings using an intelligent active alert application. The system comprises: the location module, the environment sensing module, the central-control module and the application system as well as the framework of the system as in Fig.1. Each module was controlled by the central-control module and connected with the database. The location and environment sensing modules transmitted information to the database. The central-control module sent commands to other modules according to the information from the database.
The database sent the information regarding evacuation to the application system. Each module is described below.
The Location module was used to inform the system how many people are in the space, and each one’s know location in the building. The VSLS was used to build virtual tags in the building and the On-line and Off-line system was used to determine the location of the user. The location information was transmitted to the database. And the algorithm also used a number of spaces for operating the guidance for each person.
The Environment sensing module responded to variations of in the environment. It used the ZigBee wireless sensor network to collect environment information. This module used a controller to control the speed of collection and the state of the environment. In the system, this module was used for real-time response to changes in environment and feed back to the Central-control module. The central-control module can use the data to operate the guidance, or the user can inquire about the situation of the environment via the central-control module.
The Central-control module integrated all of the modules in the system. It saved the location and environment information in the database to be used by the algorithm. The central-control
79 system also supported user inquiries regarding location and environment. Users could use their mobile device to connect to the server via the Internet at any location. The central system consists of the crowd guidance algorithm and the database system. The algorithm was started or stropped by the central-control system which also stored the results to the database. The database recorded all information and events and provided them to the other systems as needed. The framework of the central system is as given in Fig.1.
The Application system gave users their location and the information about environment.
It had two different modes: normal mode and danger mode. Each mode had a different function in providing the information and evacuation guidance. This is explained in detail in the next chapter.
Figure 1. System Framework IV. Application Design
The application system is a very important component of the system design. It was used to supply path information and guidance for each user. The user could obtain any information they needed from the application system including the temperature, humidity, illumination, location, etc.
In fact, the application is the only way to establish contact with user and system. Therefore, the correct, useful information is vital. For this purpose the two modes were designed, “normal mode” and “danger mode”, as in Fig.2. Each mode provided different information.
Figure 2. Design of Application System In normal mode, the system provided environmental information and the location of the user, user can inquire information and the location anytime and anywhere. User could also check information regarding another space even when not in the room. With danger mode, the application only provided the direction of evacuation. The system used well-marked arrow pointers to show the direction and used an Electronic Compass to keep the arrow pointing the right way. At the same time of the user evacuation, system changed the color of the arrow for a different space. The arrow would be red when user nears the fire, yellow if between fire and exit, and green if near exit. The purpose is not only giving directions, but also preempting and avoiding irrational behavior by building confidence and a sense of security.
A. Normal Mode
Should there not be any fire, the application would operate “Normal Mode”. When in normal mode, the mobile device would send the IMEI to the Server. Then it would find the Tag ID which is being mapping with the IMEI, and the server would return the location of the user and display it on the device. At the same time, the application would show the ZigBee sensors on the device.
Should a user want to get the information from a sensor just touching the icon would be sufficient.
The application would get the temperature, humidity, illumination at a sensor from the server and display it on the device. The flow chart of normal mode is as in Fig.3.
80 server. Then the direction of evacuation would be indicated using an arrow point. In case of the user facing in the wrong direction, the application would use the Electronic Compass to rotate the arrow in the right direction after recoding the angle between the building and terrestrial magnetism. At the same time, the system would determine the distance between user to exit, and indicating it with different color arrows: green, yellow, and red.
The arrow would change the color real-time when user nears the fire, nears exit, and between fire and exit. And the flow chart in case of danger mode is as Fig.4.
Figure 4. Flow Chat of Danger Mode V. Prototype Implementation In the prototype, a system with eight sensor nodes was built on one level with two exits. HTC One X was used as mobile device. It had a 4.7-inch LCD screen, quad-core 1.5GHz, 1 GB Ram, and used Wi-Fi to connect to the Internet.
The application was programmed with JAVA on Android 4.0. The two different modes in could be set on either normal or fire. In normal mode, the user could check the environment information as well as location, and could click the sensor icon to check information about each place, as in Fig.5 and Fig.6.
Figure 5. Information and Location in Normal Mode
Figure 6. Environment information of the sensor In case of an emergency, the system would send according to the user’s position as in Fig.8.
Figure 7. Device receiving a warning message
81 Figure 8. Different Color Arrows
VI. Conclusions
During an emergency, it is important that evacuees be directed efficiently. In this study, an emergency evacuation system was designed with a wireless sensor network, RFID location technology, and mobile application. It could effectively guide people along an evacuation path using a well-marked arrow when in danger mode and providing information of the environment in normal mode. Moreover, an Electronic Compass was used to make the arrow point in the right direction. In the prototype implementation, a real-time emergency evacuation system was applied to navigate towards the exit using evacuation guidance. In future, the emergency evacuation system can be linked with fire-fighting equipment or a station house, making a building safer and smarter.
Acknowledgment
This paper was supported by Architecture and Building Research Institute, Ministry of the Interior (research no. 101301070000G0020).
References
[1] Gi Heung CHOI, Gi Sang CHOI, Joo Hyoung Jang,” A Framework for Wireless Sensor Network in Web-based Monitoring and Control of Indoor Air Quality (IAQ) in Subway Stations”, Computer Science and Information Technology (ICCSIT), pp. 378-382, 2009.
[2] Yuh-Ming Cheng,” Using ZigBee and Room-Based Location Technology to Constructing an Indoor Location Based Service Platform,” IIH-MSP’09, pp.
803-806, 2009.
[3] Cheng Jen-Hsiang, Tseng Yu-Chee, Kuo Lun-Chia, Chiang Jen-Chieh, Lin Wan-Jung ,” LEGS: A Load-balancing Emergency Guiding System Based on Wireless Sensor Networks ", PERCOM Workshops, pp.486-488, 2012
[4] Z. Fang, J.P. Yuan, Y.C. Wang, S.M. Lo “Survey of pedestrian movement and development of a crowd dynamics model” Fire Safety Journal, vol.43, issue 6, pp. 459-465, 2008.
[5] Sang Guoming, Song Liwei, “The Design and Implementation of a Farmland Monitoring Wireless Sensor Network”, Second Pacific-Asia Conference on Circuits, Communications and System (PACCS), vol.1, pp.355-358, 2010.
[6] G. Jin, X. Lu, and M. S. Park, "An indoor localization mechanism using active RFID tag," IEEE International Conference on Sensor Networks, Ubiquitous, and Trustworthy Computing, pp. 4, 2006.
[7] I. Jawhar, N. Mohamed, M. Mohamed, and M. Aziz, "A Routing Protocol and Addressing Scheme for Oil, Gas, and Water Pipeline Monitoring Using Wireless Sensor Networks," The Fifth IEEE/IFIP International Conference on Wireless and Optical Communications Networks (WOCN2008), 2008.
[8] Q. Li, M. DeRosa, D. Rus, “Distributed algorithms for guiding navigation across a sensor network.”
Proceedings of Ninth Annual International Conference on Mobile Computing and Networking, September, pp.
313–326, 2003.
[9] M. G. Lee, K. M. Yu, and W. C. Lai, "Implement a RFID-Based Indoor Location Sensing System Using Virtual Signal Mechanism," 2010, pp. 168-174.
[10] L. M. Ni, Y. Liu, Y. C. Lau, and A. P. Patil,
"LANDMARC: indoor location sensing using active RFID," Wireless networks, vol. 10, pp. 701-710, 2004.
[11] S. Pu, S. Zlatanova. “Evacuation route calculation of inner buildings.” In P. van Oosterom, S. Zlatanova, and E. M. Fendel, editors, Geo-Information for Disaster Management, pp. 1143-1161, 2005.
[12] R. Tesoriero, J. Gallud, M. Lozano, and V. Penichet,
"Using active and passive RFID technology to support indoor location-aware systems," Consumer Electronics, IEEE Transactions on, vol. 54, pp. 578-583, 2008.
[13] Y. C. Tseng, M. S. Pan, and Y. Y. Tsai, "A distributed emergency navigation algorithm for wireless sensor networks," IEEE Computers, vol. 39, pp. 55-62, 2006.
[14] Tan R., Xing G., Wang J., Tan X., “Profiling Aquatic Diffusion Process Using Robotic Sensor Networks”, IEEE Transactions on Mobile Computing, Issue: 99 , pp.1, 2013.
[15] Peng Wang, Peter B. Luh, Shi-Chung Chang, , Jin Sun ,
“Modeling and optimization of crowd guidance for building emergency evacuation”, Automation Science and Engineering, pp. 328-334, 2008.
[16] K. M. Yu, J. Y. Liou, B. H. Yeh, C. W. Yu, C. C. Tien, C. H. Wang, and P. Y. Wang, "CALE: A
[16] K. M. Yu, J. Y. Liou, B. H. Yeh, C. W. Yu, C. C. Tien, C. H. Wang, and P. Y. Wang, "CALE: A