Professional Documents
Culture Documents
Help From The Sky: Leveraging Uavs For Disaster Management
Help From The Sky: Leveraging Uavs For Disaster Management
Help From The Sky: Leveraging Uavs For Disaster Management
L
arge-scale natural disasters test the and response. When a disaster occurs, the most
most fundamental human instinct important issue is preserving human lives. In
of survival by inflicting massive and this context, the first 72 hours after the disas-
often unpredictable losses of life and ter hits are the most critical, which means that
property. Various types of natural search and rescue (SAR) operations must be
disasters, such as geophysical (earthquake, tsu- conducted quickly and efficiently. The major
nami, volcano, landslide, and avalanche), hydro- problem is the lack of communication and situ-
logical (flash-floods, debris flow, and floods), ational awareness during a disaster, forcing first
climatological (extreme temperature, drought, responder teams to improvise and thus degrad-
and wildfire) and meteorologi- ing the efficiency of the rescue mission.2
cal (tropical storm, hurricane, This article reviews the latest advances in
Milan Erdelj and Enrico Natalizio sandstorm, and heavy rain- UAVs for network-assisted first response to di-
Sorbonne Universités, UTC CNRS fall), have resulted in the loss of saster management and identifies open issues
Kaushik R. Chowdhury many lives. There has also been that need to be solved. In particular, we pres-
Northeastern University, Boston an increase in material losses ent an approach for classifying disasters, and
caused by such disasters on the we outline suitable network architectures for
Ian F. Akyildiz order of 100–150 percent over effective disaster management based on these
Georgia Institute of Technology the past 30 years.1 Many efforts classifications.
are underway to recognize and
forecast the occurrence of natu- UAV-Assisted Disaster Management
ral disasters to help us react in a The response time of disaster management per-
timely manner and quickly and efficiently assess sonnel during a natural disaster is key in sav-
the damage, address the outages, and restore ing the lives of those in the affected areas. The
normalcy. most efficient situational awareness is achieved
Acknowledging the need for bolstering di- through aerial assessment—UAV networks.
saster resilience, here we describe a vision for Different regulations apply to the usage of
leveraging the latest advances in wireless sen- UAVs, depending on the country, but during
sor network (WSN) technology and unmanned a disaster, special authorizations are usually
aerial vehicles (UAVs) to enhance the ability of granted to flying devices to help first responders
network-assisted disaster prediction, assessment, assess the situation as quickly as possible.
TABLE 1
Various types of drones.
Preparedness:
WSN with limited UAV roles
Monitoring and surveying
Early Warning Systems (EWS)
Different types of wireless sensors are statically deployed in the potential disaster area
An occurrence of a disaster triggers WSN reporting with optional UAV support
Assessment: Partial
Situational awareness No WSN UAV UAV WSN No UAV
WSN
Damage assessment
Structural inspection Damage assessment and structural Damage assessment is done by WSN information fusion for
inspection is being done by UAVs UAVs, backed up by the operational situational awareness
part of WSN
• type C: meteorological (tropical storm, tions. In this case, the WSN must play to a centralized location, where the
hurricane, sandstorm, heavy rainfall). a dominant role, with partial support information is logged. Here, the sim-
made available through UAVs. plest option is to use commercial, off-
Note that type A disasters render the the-shelf cellular modem technology in
existing WSN infrastructure for moni- Stage 1: Disaster Preparedness the sensors, although this increases the
toring nonoperational. The assessment The preparedness phase doesn’t have weight and cost of the sensors. Simone
and response and recovery phases are a predefined duration and could start Frigerio and his colleagues presented a
performed mainly using UAVs. Type B several years before the anticipated di- deployment scenario of landslide moni-
disasters partially impact the existing saster event, culminating with its actual toring in the Italian Alps,10 where the
WSN infrastructure. In this case, the occurrence. For all three disaster types, WSN integrated different sensors to
role of UAVs is twofold: to reconnect the WSN plays the lead role, receiving monitor displacements caused by land-
the operational parts of WSN and to limited support from the UAV. Figure 3 slides and trigger an alarm in the case
perform other dedicated tasks. Type illustrates a case study for flood and of debris flow.
C disasters mainly focus on meteoro- landslide monitoring. Aerial surveillance via UAVs has
logical events, because the UAV can’t In the example scenario, multiple de- limited use in such types of disasters,
operate reliably during the assessment ployed sensors collect physical informa- which require ground-based measure-
phase and has limited operational use tion—the water level at the monitored ments, because the operational time of
in the disaster response and recovery bank and vibration/displacement on the UAVs might not be sufficient to cap-
phase due to unstable weather condi- the mountain side—and forward it ture the different trends in the natural
Sensor and robot network Radio network In the case of type B disasters, the
WSN infrastructure is partially opera-
tional, so it might still be used in con-
junction with the deployed UAV net-
Sensor network Radio access network work, which can serve as bridging nodes
and sustain the overall WSN topology.
Sensor-actor network architectures,
which have been studied extensively
Standby robot network Connected users elsewhere,12 can be adopted in this sce-
nario. Mobile actors—UAVs here—
might move closer to regions of network
partitions caused by loss of multiple sen-
sors and act as forwarding relays for the
WSN. Although type B disasters bring
about interesting joint roles of UAVs and
Sensor network sensors, there are additional consider-
ations. For example, the UAV can serve
as the relay node to bridge the network
Figure 3. An example mixed-WSN-UAV deployment scenario for disaster partition only for a short duration, so
preparedness for flood and landslide monitoring. Multiple sensors collect physical the comparatively long-lived WSN must
information—the water level at the monitored bank and vibration/displacement on buffer and distribute packets along the
the mountainside—and forward it to a centralized location. end-to-end chain.
Gurkan Tuna, V. Cagri Gungor, and
Kayhan Gulez have presented an inter-
parameters being sensed. Instead of When the task assignment is completely esting network paradigm in the context
sensing, UAVs can play a role by as- centralized, it’s possible to partition the of mobile robots that can also be con-
suming the load of data delivery from physical space into known regions and sidered for UAVs.13 In their work, be-
the resource-constrained sensors. For assign one or more UAVs per region. cause the WSN is still operational and
example, as shown with the “standby When the task assignment is decen- able to route packets to the remote sink,
robot network” in Figure 3, stand-by tralized, the UAVs must first establish the mobile units perform more of the
UAVs can be called into active opera- an aerial mesh that allows a fully con- exploratory tasks but then leverage the
tional service to perform the function of nected network through local coordi- WSN as the data-forwarding backhaul.
so-called data mules. nation (see Figure 4). Multiple UAV For type B disasters, we recommend
Our recommendation for this stage is stations, strategically deployed over a taking advantage of the existing WSN
to optimize WSN data acquisition and wide geographical area, can guarantee infrastructure and dedicating a part of
data analysis to assess the probability that at least some parts of the UAV in- the UAV network for WSN infrastruc-
of future disaster occurrences, using frastructure are operational, even af- ture reconnection. The WSN can not
UAVs as data mules (see Table 2). ter the disaster has occurred. Recent only acquire environmental data but
works, such as that by Marco Di Fe- also help reconnect disjoint parts of the
Stage 2: Disaster Assessment lice and his colleagues,11 rely on using UAV network.
This stage occurs when a disaster is in attraction and repulsion spring forces Given the particular nature of type
progress, rendering parts of the topo- in defining actions of UAVs, with sep- C disasters, there are instances of vio-
graphical region unusable for vehicular arate air-to-air springs (to form the lent turbulence, strong winds, and other
traffic or human habitation. The focus aerial mesh), air-to-ground springs (to weather-related artifacts that don’t al-
of the wireless network shifts from mon- connect the users), and air-to-frontier low safe airborne operation of the UAVs
itoring to providing an accurate assess- springs (to allow for the exploration of during the assessment phase. When sit-
ment of the situation. The main task here new spatial locations). uational awareness must be delegated
is surveying the land area for available Consequently, for type A disasters, to the WSN alone, a viable approach
resources and relaying this data back to we recommend using heterogeneous seems to be to use deployments such as
the control center, all in real time. UAV networks comprising fixed-wing DistressNet, an ad-hoc wireless archi-
For type A disasters, the UAVs must UAVs to scan the area and identify im- tecture that supports disaster response
form an independent network, with- portant points to be covered and sur- with distributed collaborative sensing,
out support from the ground sensors. veyed by rotary-wing UAVs. topology-aware routing using a multi-
Disaster stage
Disaster type 1. Preparedness 2. Assessment 3. Response and recovery
Type A (geophysical Optimize WSN data acquisition Use heterogeneous UAV networks Use different camera types and
or hydrological) and data analysis to assess the comprising fixed-wing UAVs to scan specialized sensors and actuators
probability of future disaster the area and identify important mounted on UAVs, dedicated for
occurrences, using UAVs as points to be covered and surveyed by rescue missions and supply delivery.
data mules. rotary-wing UAVs.
Type B Same as above Exploit the existing WSN infrastructure Maximize the data provided by
(climatological, and dedicate a part of the UAV network the WSN to improve the efficiency
hydrological, or for WSN infrastructure reconnection. of the search and rescue missions
human-induced) The WSN can acquire environmental executed by UAVs.
data and help reconnect disjointed
parts of the UAV network.
Type C Same as above Focus on the data provided by the Use the fully functional WSN to
(meteorological) WSN and other available information reconnect the impaired UAV
sources (such as social networks). networks.
monitoring the information that flows ous issues that the use of UAVs implies, the designated location, although this
back and forth from the disaster area to we have chosen the ones with the most involves higher transmission power and
the end controller caused human errors important impact on communication. increased impact on the 3D propagation
in the operation of the UAV, and nega- environment. On the other end, there
tively impacted its performance in the Type A and B Disasters is a tradeoff between the advantage of
rescue mission. Focusing on type A and B disasters, the aerial stability during handover-related
An existing WSN can also contrib- following are the issues and challenges messaging with low transmission power
ute to the on-the-fly establishment of that need to be addressed. and the correspondingly lengthy dura-
multihop wireless access networks. tion for completing the entire handover
The architecture Quang Tran Minh Creating and maintaining the informa- process.
and his colleagues have proposed17 ex- tion relay network. The relaying net-
tends Internet connectivity from sur- work formed by the UAVs is completely Type C Disaster: Strengthening
viving access points to disaster victims aerial and must have a high level of re- Hardware
through individual mobile devices. silience toward link outages owing to With a type C disaster, UAV physical
Similar concepts can be extended for motion-related changes or energy-level constraints compromise communica-
the mixed WSN-UAV architecture, changes among the UAVs. Addressing tion. In the context of disaster man-
where UAVs form the virtual access this issue requires a two-stage process: agement, one of the most important
points and the WSN connects to this an initial round of centralized determi- constraints imposed on the use of UAV
UAV network. nation of optimal relay points (which networks is their resistance to weather
Our recommendation for a type B di- we call anchors) that connect the disas- conditions. In effect, it’s reasonable to
saster is to maximize the data provided ter region to the nearest RAN, followed assume that the appearance of a natu-
by the WSN to improve the efficiency by a round of decentralized correction ral disaster is followed by other natural
of the SAR missions executed by UAVs. during deployment. calamities that would disable the use of
In a type C scenario, UAVs are lim- UAVs. Therefore, it’s important to fo-
ited in their ability to gather useful Supporting in-network data fusion. The cus on the development of specialized
information from the disaster site, but video/images collected by the UAVs hardware suitable for disaster environ-
they can operate from the periphery. present an overview of the situation. ments, as well as control algorithms
Assuming the disaster involves major However, affected humans might also that could improve the collective be-
destruction to the communications use various social media or forward havior and agility of a UAV network.
infrastructure, where cellular towers text messages and images via the UAV
or fixed base stations are rendered in- relay network. Such activity offers General Issues
effective, the only solution is for sen- fine-grained on-the-ground informa- Issues that need to be tackled regardless
sors to forward their data using low tion that can be fused at the control of the disaster type can be summed up
power, forming multihop relay chains center with the high-definition UAV in the following.
to the edge of the affected region. The feeds. Existing source/channel coding
advantage of using UAVs is that the from the domain of multimedia sen- Automating network maintenance and
pick-up point at this edge can be dy- sor networks isn’t sufficient, because UAV charging. Battery-powered UAVs
namically decided based on the sur- existing coding considers a static net- might need to intermittently dissoci-
viving elements of the initial architec- work topology with varying channel ate from the relay network for charg-
ture. The use of mobile UAV stations conditions.18 ing. Duty-cycling these UAVs—that is,
proposed in our work can ensure the selecting their alternating operational
rapid UAV deployment and prompt Addressing handover issues. Unlike and charging durations—requires
UAV network setup, thus lowering handoff in cellular systems, the hand careful optimization formulations
the response time and increasing the over among UAVs—such as during to maintain relay-path connectivity,
disaster recovery rate. recharging events—is considerably provide an adequate level of service to
Our recommendation is to use the more involved. A handover involves rep- users, and minimize the downtime of
fully functional WSN to reconnect the licating the exact operational state in the each UAV.
impaired UAV networks. incoming UAV—including forwarding Interesting problems exist in this
tables, packets in the buffer, and data space. The first is performing optimal
Open Issues and Challenges fusion rules—which escalates the mes- handoffs between the roles of survey-
Involving UAVs in disaster manage- saging between the UAVs. The hand ing, last-mile communication with
ment has several networking-related over process can begin early, during users, and data relaying. Another is
research challenges. Among the numer- the approach time of the UAV toward choosing the charging duration—that
R
work of the IMATISSE (Inundation Monitoring
and Alarm Technology in a System of SystEms) tions Upon Low-Cost Scalable Systems,”
ealizing the next-generation project, which is funded by the Region Picardie, Mobile Networks and Applications, vol.
France, through the European Regional Develop- 18, no. 5, 2013, pp. 651–663.
architectures proposed here
ment Fund (ERDF). It was supported in part by
will require creating new the US Office of Naval Research under grant no. 4. Drones for Disaster Response and Relief
network paradigms, such N00014-16-1-2651. Operations, American Red Cross, 2015;
www.issuelab.org/resources/21683/21683.pdf.
as aerial WSDNs, and enhancing es-
tablished theoretical frameworks, such References 5. D. Camara, “Cavalry to the Rescue:
as wireless sensor-actor networks. Fur- Drones Fleet to Help Rescuers Opera-
thermore, we need to design reliable and 1. “Loss Events Worldwide 1980–2014,” tions over Disasters Scenarios,” Proc.
map, NatCatSERVICE, 2015; www. 2014
IEEE Conf. Antenna Measure-
effective networks of UAVs to minimize munichre.com/site/touch-naturalhaz- ments & Applications (CAMA), 2014,
loss of life and property. We hope our ards/get/documents_E2080665585/mr/ pp. 16–19.
6. M. Erdelj and E. Natalizio, “UAV-Assisted 11. M. Di Felice et al., “Self-Organizing 2015, pp. 1–18; doi: http://dx.doi.
Disaster Management: Applications and Aerial Mesh Networks for Emergency org/10.1016/j.comnet.2015.05.007.
Open Issues,” Proc. 2015 Int’l Conf. Communication,” IEEE Symp. Personal,
Computing, Networking and Communi- Indoor, and Mobile Radio Communi- 16. G.-J. Kruijff et al., “Rescue Robots at
cations (ICNC), 2016, pp. 1–5. cation (PIMRC), 2014; doi: 10.1109/ Earthquake-Hit Mirandola, Italy: A Field
PIMRC.2014.7136429. Report,” Proc. 2012 IEEE Symp. Safety,
7. A. Gaszczak, T.P. Breckon, and J. Han, Security, and Rescue Robotics (SSRR),
“Real-Time People and Vehicle Detection 12. T. Melodia et al., “Communication and 2012; doi: 10.1109/SSRR.2012.6523866.
from UAV Imagery,” Proc. SPIE, 2011; Coordination in Wireless Sensor and Actor
doi: 10.1117/12.876663. Networks” IEEE Trans. Mobile Comput- 17. Q.T. Minh et al., “On-the-Fly Estab-
ing, vol. 6, no. 10, 2007, pp. 1116–1129. lishment of Multihop Wireless Access
8. K. Fujii, K. Higuchi, and J. Rekimoto, Networks for Disaster Recovery,” IEEE
“Endless Flyer: A Continuous Flying Drone 13. G. Tuna, V.C. Gungor, and K. Gulez, Comm. Magazine, vol. 52, no. 10, 2014,
with Automatic Battery Replacement,” “An Autonomous Wireless Sensor Net- pp. 60–66.
Proc. IEEE Conf. Ubiquitous Intelligence work Deployment System Using Mobile
and Computing and Autonomic and Robots for Human Existence Detection 18. T. Banerjee, K.R. Chowdhury, and D.P.
Trusted Computing (UIC/ATC), 2013, in Case of Disasters,” Ad Hoc Networks, Agrawal, “Using Polynomial Regression
pp. 216–223. Feb. 2014, pp. 54–68. for Data Representation in Wireless Sensor
Networks,” Wiley Int’l J. Communication
9. E. Montijano et al., “Vision-Based Dis- 14. S. George et al., “DistressNet: A Wire- Systems, vol. 20, no. 7, 2007, pp. 829–856.
tributed Formation Control without an less Ad-Hoc and Sensor Network Archi-
External Positioning System,” IEEE Trans. tecture for Situation Management in
Robotics, vol. 32, no. 2, 2016, pp. 339–351. Disaster Response,” IEEE Communi-
cations Magazine, vol. 48, no. 3, 2010,
10. S. Frigerio et al., “A Web-based Platform pp. 128–136. Read your subscriptions
for Automatic and Continuous Landslide through the myCS
Monitoring: The Rotolon (Eastern Ital- 15. I.F. Akyildiz et al., “SoftAir: A Software publications portal at
ian Alps) Case Study,” Computers and Defined Networking Architecture for 5G http://mycs.computer.org.
Geosciences, vol. 63, 2014, pp. 96–105. Wireless Systems,” Computer Networks,
EXECUTIVE STAFF
Executive Director: Angela R. Burgess; Director, Governance & Associate
Executive Director: Anne Marie Kelly; Director, Finance & Accounting:
PURPOSE: The IEEE Computer Society is the world’s largest association Sunny Hwang; Director, Information Technology & Services: Sumit
of computing professionals and is the leading provider of technical Kacker; Director, Membership Development: Eric Berkowitz; Director,
information in the field. Products & Services: Evan M. Butterfield; Director, Sales & Marketing:
MEMBERSHIP: Members receive the monthly magazine Computer, Chris Jensen
discounts, and opportunities to serve (all activities are led by volunteer
members). Membership is open to all IEEE members, affiliate society COMPUTER SOCIETY OFFICES
members, and others interested in the computer field. Washington, D.C.: 2001 L St., Ste. 700, Washington, D.C. 20036-4928
OMBUDSMAN: Email ombudsman@computer.org. Phone: +1 202 371 0101 • Fax: +1 202 728 9614 • Email: hq.ofc@computer.org
COMPUTER SOCIETY WEBSITE: www.computer.org Los Alamitos: 10662 Los Vaqueros Circle, Los Alamitos, CA 90720
Phone: +1 714 821 8380 • Email: help@computer.org
Next Board Meeting: 30 January–3 February 2017, Anaheim, CA,
USA MEMBERSHIP & PUBLICATION ORDERS
Phone: +1 800 272 6657 • Fax: +1 714 821 4641 • Email: help@computer.org
EXECUTIVE COMMITTEE Asia/Pacific: Watanabe Building, 1-4-2 Minami-Aoyama, Minato-ku, Tokyo
President: Jean-Luc Gaudiot; President-Elect: Hironori Kasahara; Past 107-0062, Japan • Phone: +81 3 3408 3118 • Fax: +81 3 3408 3553 •
President: Roger U. Fujii; Secretary: Forrest Shull; First VP, Treasurer: Email: tokyo.ofc@computer.org
David Lomet; Second VP, Publications: Gregory T. Byrd; VP, Member &
Geographic Activities: Cecilia Metra; VP, Professional & Educational IEEE BOARD OF DIRECTORS
Activities: Andy T. Chen; VP, Standards Activities: Jon Rosdahl; VP, President & CEO: Karen Bartleson; President-Elect: James Jefferies; Past
Technical & Conference Activities: Hausi A. Müller; 2017–2018 IEEE President: Barry L. Shoop; Secretary: William Walsh; Treasurer: John
Director & Delegate Division VIII: Dejan S. Milojičić; 2016–2017 IEEE W. Walz; Director & President, IEEE-USA: Karen Pedersen; Director &
Director & Delegate Division V: Harold Javid; 2017 IEEE Director-Elect & President, Standards Association: Forrest Don Wright; Director & VP,
Delegate Division V-Elect: John W. Walz Educational Activities: S.K. Ramesh; Director & VP, Membership and
Geographic Activities: Mary Ellen Randall; Director & VP, Publication
BOARD OF GOVERNORS Services and Products: Samir El-Ghazaly; Director & VP, Technical
Term Expiring 2017: Alfredo Benso, Sy-Yen Kuo, Ming C. Lin, Fabrizio Activities: Marina Ruggieri; Director & Delegate Division V: Harold Javid;
Lombardi, Hausi A. Müller, Dimitrios Serpanos, Forrest J. Shull Director & Delegate Division VIII: Dejan S. Milojičić
Term Expiring 2018: Ann DeMarle, Fred Douglis, Vladimir Getov, Bruce
M. McMillin, Cecilia Metra, Kunio Uchiyama, Stefano Zanero
Term Expiring 2019: Saurabh Bagchi, Leila De Floriani, David S. Ebert, Jill
I. Gostin, William Gropp, Sumi Helal, Avi Mendelson revised 2 Dec. 2016