Overview of Research at CERT

Prof. Sharad Mehrotra

Donald Bren School of Information and Computer Sciences

Director, Center for Emergency Response Technologies

Nov 4th, 2009

EMWS 2009

Center for Emergency Response Technologies

• CERT Goals– lead research, technology development & coordination of ongoing

projects on role of IT to improving emergency response. – provide a forum for collaboration between academia, industry, and

government agencies.

• 20+ associated faculty, 40+ students, staff, programmers, researchers– Computer Science, Social Science, Engineering

• Close partnerships with local & state agencies, industry.– City of LA, city of SD, OCFA, LA Fire, …

• Funded by grants from NSF and DHS– Including ~$15 million award for Project RESCUE/Responsphere

2

EMWS 2009

CERT: Faculty Involved

• Naveen Ashish• Scott Bartell• Carter Butts• Marlon Boarnet• Mike Carey

• Nikil Dutt

• Rina Dechter• Magda El Zarki• Rufus Edwards• Alex Ihler

• Ramesh Jain

• Steve Jenks

• Jay Jayakrishnan• Dmitri Kalashnikov• Chen Li• Aditi Majumder• Gloria Mark• Gopi Meenakshisundaram• Sharad Mehrotra, Director• Bonnie Nardi• Deva Ramanan• Walt Scacchi• Padhraic Smyth• Nalini Venkatasubramanian

Multidisciplinary faculty from computer science, engineering, social sciences, statistics and medicine

EMWS 2009

CERT Partners

Industrial Partners5G Wireless

Broad-ranged IEEE 802.11 networking

AMDCompute Servers

Apani NetworksData security at layer 2

Asvaco1st responder (LAPD), and threat

analysis software

BoeingCommunity Advisory Board

Member

CanonVisualization equipment SDK

ConveraSoftware partnership

Cox CommunicationsBroadcast video delivery

D-LinkCamera Equipment and SDK

Ether2Next-generation ethernet

IBMSmart Surveillance Software (S3)

and 22 e330 xSeries servers

ImageCat, Inc.GIS loss estimation in emergency

response

MicrosoftSoftware

PrintronixRFID Technology

The School Broadcasting Company

School based dissemination

Vital Data TechnologySoftware partnership

Walker WirelessPeople-counting technology

Government PartnersCalifornia Governor’s Office of Emergency

Services

California Governor’s

Office of Homeland Security

City of Champaign City of Dana Point

City of Irvine City of Los Angeles

City of Ontario

Fire DepartmentCity of San Diego

Department of Health and Human Services – Centers

for Disease Control

Lawrence Livermore

National Laboratory

Los Angeles CountyNational Science

Foundation

Orange CountyOrange County Fire

Authority

U.S. Department of

Homeland Security

EMWS 2009

CERT Vision

Right Information to the Right Person at the Right Time

can result in dramatically better response

Response Effectiveness• lives & property saved • damage prevented• cascades avoided

Quality & Timeliness of

Information

Situational Awareness• incidences• resources• victims• needs

Quality of Decisions• first responders• consequence planners• public

EMWS 2009

Response Planning

Events, people, resources

Sensors create awareness Awareness enhances Response

Operational planning

@ crisis site

Consequence Analysis @ EOC

Multimodal Situational Awareness in Response

Situational Awareness

Transportation Infrastructure

Control actions

Evacuation planning, dangerous routes, occupancy analysis, exposure analysis

Location of victims, hazardous material, status of fire fighters

Physical World

MULTI-MODAL SENSORS

Incident level Response

EMWS 2009

CERT Structure

Core research driving new technologies

IT systems & tools of direct relevance to crisis response

Simulations/Drills that mimic crisis response activities

System Artifacts

Societal Impact - Crisis Response

Testbeds

FundamentalResearch

TRANSPORTATIONTRANSPORTATION

RESPONSPHERERESPONSPHERE

GLQ GLQ

CHAMPAIGNCHAMPAIGN

INFORMATION SHARING

INFORMATIONCOLLECTION

INFORMATIONANALYSIS

INFORMATIONDISSEMINATION

EMWS 2009

IT Research at CERT

• Infrastructure Level– Goal:

– Resilient infrastructures for sensing, computation, communication and Data management

– Challenges/issues– Graceful performance degradation under extreme loads, catastrophic

failures, surge demands, scalable infrastructure, inexpensive, easy to deploy

• Information Level– Goal:

– Real-time Situational Awareness from multimodal inputs to support decision making

– Challenges/issues– Diversity of information sources, multiple modalities, and needs,

uncertainty in data, challenge of scale, human-as-a-sensor• Deployment and Societal Issues

– Goal: – Enable technology solutions to work in societal scale deployments

– Challenges/issues– Privacy, Novel IT practices, Cost/Culture, Reliability, Digital Divide

EMWS 2009

IT Infrastructure Research @ CERT

• Reliable Networking at the Crisis Site – Cognitive MANETS that self-adapt at all levels of protocol stack to adjust to load

and add robustness (Jafarkhani/Yousefi’zadeh)– Disruption tolerant hybrid Instant Networks that combine “store-n-forward” with

ABC networks to support quality guarantees (Venkat)

• Middleware for reliable & Scalable Sensing– SATWARE: Exploit application tolerance to errors/delays to build resilience at

higher layers (Mehrotra/Venkat)

• Reliable Mobile computing platforms– Power aware cross-layer architectures for mobile applications (Dutt/Venkat)

• Scalable Data Management– ASTERIX: Scalable management of massive semi-structured databases that

leverage parallelism (Carey/Li)

– Database as a service: reliable & secure data management in the cloud that

provides privacy/confidentiality guarantees (Mehrotra)

• Scalable Information Dissemination – Internet-scale rapid dissemination using P2P infrastructure (Mehrotra/Venkat)

EMWS 2009

Information Systems Research @ CERT

• Sensor Driven Data Analysis– Anomaly detection & event detection in sensor streams (Smyth/Ihler) – Multi-sensor framework for localization (Ihler/Mehrotra/Venkat)– Sensor-driven occupancy analysis & forecasting (Dechter/Smyth)

• Multimodal Event Detection– From conversational speech (Kalashnikov/Mehrotra/Venkat)– From textual input (Ashish/Mehrotra)– From Video/Multimodal Data (Jain/Mehrotra/Ramanan)

• Event-based Situational Modeling – Event representation, event based query languages, visualization

(Jain/Mehrotra) – Uncertainty representation and probabilistic query processing

(Ashish/Dechter/Smyth)

• Multi-sensor Data Cleaning– GDF: a general purpose toolkit for reducing uncertainty/improving data

quality in event databases (Kalashnikov/Mehrotra)

EMWS 2009

Societal Scale Deployment Research @ CERT

• Understanding societal use of technology in disasters– Project REACT: use of ubiquitous technologies to respond to

disruptions (Mark)– Social media & its role in public awareness / crisis communication

(Butts/Tierney/Sutton/Venkat)– Instant messaging as a communication modality (Jain/Sutton)

• Modeling societal information needs & characteristics during disasters– Modeling Information diffusion through social networks (Butts)– Information reliability from new information modalities (E.g.,

blogs, twitter) (Butts)• Challenges in technology adoption

– Cost & cultural barriers to information sharing (Sutton/Tierney)– Privacy and confidentiality in information sharing and data

collection (Mehrotra/Venkat)

EMWS 2009

CERT Projects

• Multiple interdisciplinary projects with other academic, and industrial partners

– RESCUE – a collaboration of CERT with 6 other research organizations/universities supported by NSF through its large ITR program

– SAFIRE – a collaboration of CERT with Imagecat supported by DHS– Next generation Alternate 911 Systems – a collaboration of CERT with

SRI supported by NSF– FireTrack Project – a collaboration with Department of epidemiology and

Deltin corporation (currently under review)– Speech-based situational Awareness – a collaboration with SRI

(currently under review)

EMWS 2009

RESCUE Project

The mission of RESCUE is to enhance the ability of emergency response organizations to rapidly

adapt and reconfigure crisis response by empowering first responders with access to accurate

& actionable evolving situational awareness

• Privacy• Security• Trust

• Natural Hazards Center• Social Science

• Data Management• Security and Trust

• Disaster Analysis • Earthquake Engineering• GIS

• Civil Engineering• Data Analysis & Mining• Data Management• Middleware & Distributed Systems

• Civil Engineering• Transportation Engineering

• Computer Vision• Networking• Multimodal Speech

Research Team

• Transporation Modeling• Urban Planning

• Privacy• Social Science • Transportation Science

• Wireless

Funded by NSF through its large ITR program

1. Extreme Networking at the Crisis Site

2. SAMI – Situational Awareness through Multimodal Input

3. Policy-Based Information Sharing Architecture

4. Customized Dissemination in the Large

5. Privacy Issues in Deployment

6. MetaSIM – An architecture for simulator integration

EMWS 2009

Situational Awareness for Fire Fighters (partnership with ImageCat)

EMWS 2009

Speech-Based Situational Awareness (collaboration with SRI)

Type of Acoustic Analysis• Human Speech: Who spoke to whom about what from

where and when• Ambient Sounds: explosions, loud sounds, screaming,

etc• Physiological Events: cough, gag, excited state of

speaker, slurring, …• Other features: too loud, too quiet for too long, …

15

Speech

Voice

Amb. Noise

Processing

Conversation Monitoring & Playback

Image & Video Tagging

Acoustic Capture Acoustic Analysis SA Applications

Spatial Messaging

Localization via Speech

Alerts

Motivation•Radio conversation is the most common communication mechanism during crisis response amongst responders

•Automatic conversation analysis can significantly enhance incident awareness leading to better response

EMWS 2009

FireTrack: Firefighter Exposure Analysis (collaboration with Deltin Corp. and Dept. of Epidemiology )

16

Goal: Reduce toxic exposures to occupational group - firefighters

•System to combine particulate/CO monitoring, physiological sensing with a framework for collecting, storing, analyzing and alerting

•Pilot studies with the LA County Fire to evaluate the efficacy in structural overhaul and in wild land fires

•Longitudinal exposure studies associated with specific groups, work tasks and activities

The WarnWatch System

Firetrack

EMWS 2009

CERT Testbeds

MetaSim Simulator (Regional Response)An integrated plug-and-play micro/macro simulation environment for diverse crisis response situations. Built in hooks for technology validation

Responsphere, UCI(incident-level response)

Different testbeds model information

flow conditions under diverse types of crisis situations

Different testbeds model information

flow conditions under diverse types of crisis situations

Responsphere(incident response) A Campus-wide infrastructure to instrument, monitor, disaster drills & to validate technologies

EMWS 2009

Responsphere Enables Drills & Technology Evaluation

• Technology Testing Exercise: 16 SEP 08– Bren Hall Evacuation

w/Campus Police Department & UCI Zone Crew 3

• Live Burn with OCFA,LA Fire and Anaheim Fire– Testing Sensing (human

bio-sensing) data collection & 2nd generation FICB

• SAFIRE / FICB Usability Study – 15 MAY 09– Freeze points identified as

critical junction / decision points to assess SA with and without FICB

18

EMWS 2009

CERT Artifacts

• Derivative research products of direct relevance to first responders– Mechanism to focus CERT research

– Opportunity for technology transfer

• Some of CERT artifacts are in current deployment at partner sites

Disaster Portal in use by City of Ontario since Sept.

2007 to improve communication of relevant

situational awareness information between first responder and the public

EMWS 2009

Closing Remarks

• CERT Outcomes– Research

– Robust disruption tolerant IT infrastructures– SA technologies for transformational

improvements to the response process by improving decision making

– Societal level technology adoption

– Testbeds – Simulation and instrumented real environments

for technology testing

– Partnerships and collaborations – industry and government partnerships

EMWS 2009

Closing Remarks

• Future plans for CERT

a)Transitioning current and ongoing CERT research to first responder– Technology transfer through CERT artifacts such as

disaster portal

b)Focus research on new opportunities to improve response by exploiting the emerging IT revolution– Cloud computing, database as a service, social

computing, peer oriented computing, grid computing, agent-based technologies, sensor networks

c) Expand research focus of CERT to beyond crisis response– Societal level adaptive systems, critical Infrastructure

protection– Example: Smart Grid

Emergency Response & Smart Grid

• Two Perspectives:

– Smart Grid as a critical infrastructure – Disruption, damage to generation & distribution systems due

to Earthquakes, fires, etc. – Opportunity to leverage vast technology base developed for

crisis response – collection, analysis, sharing, dissemination

– Grid as a critical resource– Data collected by smart grid could be valuable for response

planning– Electricity is a vital resource on which other infrastructures

depend– Hospitals, shelters, special needs, industry, ..

– Ability to control electric usage could be an important tool in effective response planning

22

Response Planning

Events, people, resources

Emergency Response & Smart Grid

Situational Awareness

Response ActionsPhysical

World

Emergency Response

Situational Awareness

Distributed Grid Control

Smart Grid

Grid Control Actions

Electric Power NeedsHospitals, shelters, specialequipment

State of the grid, outages, damage to distribution system

Grid Sensors, Smart Meters