Your SlideShare is downloading. ×
0
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Pratik Desai Ph.D dissertation defense
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×
Saving this for later? Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime – even offline.
Text the download link to your phone
Standard text messaging rates apply

Pratik Desai Ph.D dissertation defense

432

Published on

Recently, the domain of cyber-physical systems (CPSs) has emerged as a successor to the traditional embedded systems and the wireless sensor networks. The relatively new cyber-physical domain offers …

Recently, the domain of cyber-physical systems (CPSs) has emerged as a successor to the traditional embedded systems and the wireless sensor networks. The relatively new cyber-physical domain offers tight integration of control, communication and computation components to develop advanced web based application in various heterogeneous domains such as health care, disaster management, automation and environment monitoring. The applications of indoor CPSs include remote patient monitoring, smart home, etc. with focus on situation awareness via event identification from context information. The principal challenges associated with the development of situation awareness applications include uncertainty in contextual data, incomplete domain knowledge, interoperability between interconnected systems and effective utilization of spatial information.
This dissertation addresses these challenges by providing a comprehensive situation awareness framework for event comprehension utilizing raw sensor data and spatial information. Semantic web based annotation and mapping techniques are used to provide interoperability. The framework contains contextual situation awareness and location awareness stages towards achieving effective event assessment. The contextual situation awareness stage provides fuzzy abductive reasoning based architecture to transform raw physical sensor data to low-level fuzzy abstraction. These abstractions are used for event assessment with associated degree of certainty. The location awareness stage includes methodologies to hierarchically map indoor objects and define the object-event relationship in ontology, which is further exploited for event discrimination. This dissertation also presents a fusion based indoor positioning algorithm to provide accurate spatial information to assist location awareness. The algorithm uses extensive training of received signal strength (RSS) and time difference of arrival (TDoA) signals to estimate distance and position. The comprehensive framework is evaluated through an implementation of simulated indoor fire in a controlled environment.

Published in: Education, Business
0 Comments
2 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total Views
432
On Slideshare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
20
Comments
0
Likes
2
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide
  • Indoor locationInteroperabilityHelp in decision making
  • How and why location is different than environmental contextSet of related concepts, range of observation or reoccurring patterns.Derived from domain specific background knowledge for the specific situation.In CPS, perception abstraction are derived from observable sensor measurements.Example:For observable property temperature,Temperature_HighTemperature_Loware perception abstractions.Qualities, Entities
  • Cory at Kno.e.sisEntity extractionAbductive reasoning approachCrisp abstractions
  • In case of complex system and sensor failures, abductive approach provide explanations while deduction process couldn’t complete due to lack of complete rulesExample with CO2 where it failsNeed for an approach to handle fuzzy ness
  • Assumptions
  • Indoor locationInteroperabilityHelp in decision making
  • Semantic web provide common framework that alows data to be shared and reused across applications
  • Indoor locationInteroperabilityHelp in decision making
  • Transcript

    • 1. Department ofElectrical EngineeringA Semantic SituationAwareness Frameworkfor Indoor Cyber-Physical SystemsDissertation CommitteeDirector Dr. Kuldip RattanCo-Director Dr. Amit ShethDr. Marian KazimierczukDr. Frank ZhangDr. Guru SubramanyamPh.D. in Engineering Dissertation DefensePratikkumar DesaiMonday, 4/29/2013
    • 2. Department ofElectrical EngineeringEmbedded systemse.g. thermostatNetworked embeddedsysteme.g. wirelesssensor networksCyber-physicalsysteme.g. Intelligent trafficmanagementsystems
    • 3. Department ofElectrical EngineeringCyber-Physical Systemshttp://www.cs.binghamton.edu/~tzhu/Cyber : Computation, communication, and control that are discrete, logical, andswitched.Physical : Natural and human-made systems governed by the laws of physicsand operating in continuous time.Cyber-Physical Systems (CPS) : Systems in which the cyber and physicalsystems are tightly integrated at all scales and levels
    • 4. Department ofElectrical EngineeringCPS ExamplesDisasterManagementSmartGridSmartHomeRemote PatientMonitoringTrafficManagementAir TrafficControlMilitary Drones
    • 5. Department ofElectrical Engineering
    • 6. Department ofElectrical EngineeringMotivationMobile sensing platformSituation: Actual fireat chair
    • 7. Department ofElectrical EngineeringMotivationMobile sensing platformEvent : Firefrom temperature and CO2dataEvent : Firefrom temperature and CO2data
    • 8. Department ofElectrical EngineeringMotivationMobile sensing platformUncertainty: Sensor datae.g. Due to resoultion, calibrationor robustness of sensors
    • 9. Department ofElectrical EngineeringMotivationMobile sensing platformIncomplete domain knowledgee.g. Unknown sources in the environment
    • 10. Department ofElectrical Engineering
    • 11. Department ofElectrical EngineeringContextEnvironmentalcontextLocatione.g. temperature,CO2, heart ratee.g. coordinatesContextualsituationawarenessLocationawarenessContextual situation awareness:“is a process of comprehending meaningof environmental context in terms ofevents or entities”Location awareness:“is a process of identifying objects from rawspatial information and their relationshipwith the ongoing events”Context“is a physical phenomenon, measuredusing sensors, and product of an event”
    • 12. Department ofElectrical EngineeringContextual situation awareness + Location awarenessRawenvironmentalsensor dataRaw spatialinformationEntities (High levelabstractions)Object-EntityrelationshipsSituation
    • 13. Department ofElectrical Engineering
    • 14. Department ofElectrical EngineeringIntellegO0100CtemperatureCO201000ppmFireDryIceRoomHeaterNormalConditionQuality-type EntityQualityLowTempHighTempLowCO2HighCO2• Abductive reasoning• Crisp abstractionshttp://wiki.knoesis.org/index.php/IntellegoDomain KnowledgeBase
    • 15. Department ofElectrical EngineeringTemperature: 500 C HighTempCO2: 1010 ppm HighCO2Temperature: 500 C HighTempCO2: 999 ppm LowCO2DryIceRoomHeater
    • 16. Department ofElectrical EngineeringMotivationMobile sensing platformUncertainty: Sensor datae.g. Due to limitation, calibrationor robustness of sensorsIncomplete domain knowledgee.g. Unknown sources in the environment
    • 17. Department ofElectrical EngineeringFuzzy abstractions01 LowCO2ppm800HighCO212001160 ppm0.90.1µa
    • 18. Department ofElectrical EngineeringFuzzy abductive reasoning500 C1160 ppm080120CtemperatureCO208001200ppmLowTemp FireDryIceRoomHeaterNormalConditionQuality-type EntityQualityHighTempLowCO2HighCO2
    • 19. Department ofElectrical Engineering
    • 20. Department ofElectrical EngineeringEvaluation – Contextual Situation AwarenessReasoning approach Accuracy Precision RecallCrisp abductive reasoning 86 % 78.57 % 73.33 %Fuzzy abductive reasoning 94 % 92.85 % 86.66 %
    • 21. Department ofElectrical Engineering
    • 22. Department ofElectrical EngineeringSemantic Web• Semantic web:• Formally define the meaning of information on web.• Provide expressive representation, formal analysis of resources.• Ontology• Formally represents knowledge as a set of concepts withina domain and the relationships between pairs of concepts.• RDF (Resource Description Framework)• Graph-based language for modeling of information.• Allows linking of data through named properties.http://www-ksl.stanford.edu/kst/what-is-an-ontology.htmlSubject ObjectPredicateHighTempInheresInFire
    • 23. Department ofElectrical EngineeringRaw SensorDataSSNannotatedObservationsLow-levelfuzzyabstractions(qualities)FuzzyreasoningHigh-levelAbstractions(entity)SSNOntologyDomain Ontology Fuzzy Inferencerules ontologyFuzzificationRulesObservation process Perception processContextual situation awareness (Semantic modeling)
    • 24. Department ofElectrical Engineering
    • 25. Department ofElectrical Engineering
    • 26. Department ofElectrical EngineeringTraditional Indoor Localization Techniques• Active Badge and Active Bat system.• RADAR: An In-building RF-based user locationand tracking system.• RFID radar• Object tracking with multiple cameras• Computer vision based localization• Wireless Sensor NetworkRFCameraTDoA
    • 27. Department ofElectrical EngineeringTDoA (Time Difference of Arrival)ListenerBeaconStage 1: Beacon transmits RF andUs signals togetherStage 2: Listener receivesRF signal first at Trf andstarts the clockStage 3: At Tus time US signals is receivedby ListenerStage 3: Distance is calculated using ∆Tand speed of signals
    • 28. Department ofElectrical EngineeringTrilaterationNumber of nodes = 3.Outlier rejection and Multilateration
    • 29. Department ofElectrical EngineeringThe Proposed Algorithm• Utilizes fusion of RSS (received signal strength) of RFsignal and TDoA data for accurate distance estimation.• The algorithm stages:-• RSSI data training• Distance estimation• Localization• Uses TDoA as a primary distance estimation technique.• RSSI data is trained and converted into appropriatedistance measurements.• The proposed algorithm can be used in absence of one ormany TDoA links.
    • 30. Department ofElectrical EngineeringInitial Conditions• Distances between all beacons are known and fixed
    • 31. Department ofElectrical Engineering0 ? ? ?? 0 ? ?? ? 0 ?? ? ? 0LB4B2B3B10 ? ? ?? 0 ? ?? ? 0 ?R14 ? ? 00 ? ? ?R12 0 ? ?? ? 0 ?? ? ? 00 ? ? ?? 0 ? ?R13 ? 0 ?? ? ? 00 ? ? ? R1L T1L? 0 ? ? ? ?? ? 0 ? ? ?? ? ? 0 ? ?RSSI LinkTDoA LinkBeacon B1 Transmit Data
    • 32. Department ofElectrical Engineering0 R21 ? ?R12 0 ? ?? ? 0 ?? ? ? 0LB4B2B3B10 ? ? ?R12 0 ? ?? ? 0 ?R14 R24 ? 00 ? ? ?R12 0 ? ?? ? 0 ?? ? ? 00 ? ? ?R12 0 ? ?R13 R23 0 ?? ? ? 00 ? ? ? R1L T1LR12 0 ? ? R2L T2L? ? 0 ? ? ?? ? ? 0 ? ?RSSI LinkTDoA LinkBeacon B2 Transmit Data
    • 33. Department ofElectrical Engineering0 R21 R31 ?R12 0 ? ?R13 R23 0 ?? ? ? 0LB4B2B3B10 ? ? ?R12 0 ? ?R13 R23 0 ?R14 R24 R34 00 ? ? ?R12 0 R32 ?R13 R23 0 ?? ? ? 00 ? ? ?R12 0 ? ?R13 R23 0 ?? ? ? 00 ? ? ? R1L T1LR12 0 ? ? R2L T2LR13 R23 0 ? R3L T3L? ? ? 0 ? ?RSSI LinkTDoA LinkBeacon B3 Transmit Data
    • 34. Department ofElectrical Engineering0 R21 R31 R41R12 0 ? ?R13 R23 0 ?R14 R24 R34 0LB4B2B3B10 ? ? ?R12 0 ? ?R13 R23 0 ?R14 R24 R34 00 ? ? ?R12 0 R32 R42R13 R23 0 ?R14 R24 R34 00 ? ? ?R12 0 ? ?R13 R23 0 R43R14 R24 R34 00 ? ? ? R1L T1LR12 0 ? ? R2L T2LR13 R23 0 ? R3L T3LR14 R24 R34 0 R4L T4LRSSI LinkTDoA LinkBeacon B4 Transmit Data
    • 35. Department ofElectrical EngineeringEvaluation– Proposed Algorithm
    • 36. Department ofElectrical Engineering
    • 37. Department ofElectrical EngineeringBed-1Chair-2Chair-1Sofa-1Fireplace-1Treadmill-1Desk-1Stove-1Bedroom-1 Bedroom-2Gym-1Drawingroom-1Kitchen-1Plant-1IndoorEnvironment
    • 38. Department ofElectrical EngineeringHierarchical mapping ofthe indoor environment
    • 39. Department ofElectrical EngineeringChair-1Drawingroom-1xsd:floatxsd:floatxsd:floatxsd:floatxsd:floatxsd:floatxsd:stringChairPOIPOIDrawingroomis-ahas individualis-aStructuralComponenthas individualDefiningCoveragespace(Chair-1)
    • 40. Department ofElectrical EngineeringRaw location : ( x, y) = (190 cm, 570 cm)
    • 41. Department ofElectrical EngineeringObject-entity relationshipStructural Components Point of Interests EntitieshasIndividualhasApplicableEntity
    • 42. Department ofElectrical EngineeringEvaluation – Location AwarenessChair-1Sofa-1Fireplace-1Drawingroom-1Plant-1(140,560)(140,640) (430,640)(430,560)(60,480) (200,480)(60,140) (200,140)(180,110)(180,10)(110,340)(10,340)(610,360) (760,360)(610,240) (760,240)( )( )Mobile-robot route
    • 43. Department ofElectrical EngineeringReasoningapproachPrecision RecallCrisp abductivereasoning87.5 % 43.75 %Fuzzy abductivereasoning100 % 50 %Location aidedfuzzy abductivereasoning100 % 88.89 %Location aidedreasoningLocation independentreasoningFireplace
    • 44. Department ofElectrical Engineering
    • 45. Department ofElectrical EngineeringEntitiesQualitiesIndoorPositioningSystemEnvironmentSensorsSituationFuzzyAbductiveReasoningFuzzyAbstractionRulesContextual Situation AwarenessSemanticObjectIdentifierSpatialReasoningLocationAwarenessDomainKnowledgeComprehensive Framework(System level)
    • 46. Department ofElectrical EngineeringRawPhysicalContextDataSSNannotatedObservationsLow-levelFuzzyAbstractions(Qualities)High-levelAbstractions(Entities)OptimizedSituationRawLocationDataSemanticLocationIdentifierSSNOntologyDomain Ontology Fuzzy AbductiveReasoning RulesIndoor LocationOntologyComprehensive Framework(Semantic modeling)
    • 47. Department ofElectrical EngineeringTemp: 150 CCO2:1120 ppmTemp: 180 CCO2:1160 ppmLocation (b):(400,150,30)Location (a):(200,250,20)Object coverage areaMobile robot path
    • 48. Department ofElectrical EngineeringFire: 0.80Heater: 0.20Fire: 0.90Heater: 0.10Location (b):(400,150,30)Location (a):(200,250,20)Object coverage areaMobile robot path
    • 49. Department ofElectrical EngineeringFire: 0.80Heater: 0.20Fire: 0.90Heater: 0.10Location (b): ChairLocation (a): FireplaceObject coverage areaMobile robot path
    • 50. Department ofElectrical EngineeringFire: 0Heater: 0Fire: 0.90Heater: 0.10Location (b): ChairLocation (a): FireplaceObject coverage areaMobile robot path
    • 51. Department ofElectrical EngineeringKey Contributions• Developed a fusion based indoor localization algorithm toachieve accurate spatial information of the sensingplatform.• Accurate indoor localization algorithm.• Surveillance and tracking of mobile robots in indoor environments.• Integration of indoor positioning results with virtual world environment.Related papers:• P. Desai, N. Baine, and K. S. Rattan, “Fusion of RSSI and TDoA Measurements from Wireless Sensor Networkfor Robust and Accurate Indoor Localization,” in International Technical Meeting of The Institute of Navigation,2011, pp. 223–230.• P. Desai, N. Baine, and K. S. Rattan, “Indoor localization for global information service using acoustic wirelesssensor network,” in Proceedings of SPIE, 2011, vol. 8053, no. 1, pp. 805304–805304–10.• P. Desai and K. S. Rattan, “System Level Approach for Surveillance Using Wireless Sensor Networks and PTZCamera,” in 2008 IEEE National Aerospace and Electronics Conference, 2008, pp. 353–357.• P. Desai and K. S. Rattan, “Indoor localization and surveillance using wireless sensor network and Pan/Tiltcamera,” in Proceedings of the IEEE 2009 National Aerospace Electronics Conference NAECON, 2009, pp. 1–6.• An invited journal paper in preparation.
    • 52. Department ofElectrical EngineeringKey Contributions• Introduced fuzzy abstraction and inference technique tocomprehend events via handling the uncertainty in thecontext information & the ambiguity in the domainknowledge.• P. Desai, C. Henson, P. Anatharam, and A. Sheth, “SECURE: Semantics EmpoweredresCUe Environment (Demonstration Paper),” in 4th International Workshop on SemanticSensor Networks (SSN 2011), 2011, pp. 110–113.• A journal paper in preparation.• Developed semantic mapping technique for indoor objectsto aid the situational context awareness results via furtherdiscriminating not applicable events.• Developed and deployed a comprehensive situationawareness framework for cyber-physical system.• A journal paper in preparation.
    • 53. Department ofElectrical EngineeringFuture work• Richer spatio-temporal relation modeling between indoorobjects and entities• Efficient coverage space for the indoor objects• Accurate indoor localization via smartphones
    • 54. Department ofElectrical EngineeringAcknowledgements
    • 55. Department ofElectrical EngineeringQuestions?

    ×