Issues and challenges of knowledge representation and reasoning methods in situation assessment (Level 2 Fusion)

Please download to get full document.

View again

of 19
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Document Description
abstract Situation assessment (SA) involves deriving relations among entities, eg, the aggregation of object states (ie classification and location). While SA has been recognized in the information fusion and human factors literature, there still
Document Share
Document Tags
Document Transcript
  Issues and Challengesin Situation Assessment (Level 2 Fusion) ERIK BLASCHIVAN KADARJOHN SALERNOMIECZYSLAW M. KOKARSUBRATA DASGERALD M. POWELLDANIEL D. CORKILLENRIQUE H. RUSPINISituation assessment (SA) involves deriving relations among en-tities, e.g., the aggregation of object states (i.e., classification andlocation). While SA has been recognized in the information fusionand human factors literature, therestill exist open questions regard-ing knowledge representation and reasoning methods to afford SA.For instance, while lots of data is collected over a region of interest,how does this information get presented to an attention constraineduser? The information overload can deteriorate cognitive reason-ing so a pragmatic solution to knowledge representation is neededfor effective and efficient situation understanding. In this paper,we present issues associated with Level 2 Information Fusion (Sit-uation Assessment) including: (1) user perception and perceptualreasoning representation, (2) knowledge discovery process models,(3) procedural versus logical reasoning about relationships, (4) user-fusion interaction through performance metrics, and (5) syntacticand semantic representations. While a definitive conclusion is notthe aim of the paper, many critical issues are proposed in order tocharacterize future successful strategies for knowledge representa-tion, presentation, and reasoning for situation assessment. Manuscript received May 8, 2006; revised Oct. 7, 2006; released forpublication Nov. 28, 2006.Refereeing of this contribution was handled by Dr. Chee-Yee Chong.Authors’ addresses: E. Blasch, AFRL, Dayton, OH; I. Kadar, Interlink Systems Sciences, Inc.; J. Salerno, AFRL, Rome, NY; M. M. Kokar,Northeastern University, Boston, MA; S. Das, Charles River Analyt-ics, Cambridge, MA; G. M. Powell, U.S. Army RDECOM CERDECI2WD, Ft. Monmouth, NJ; D. D. Corkill, Univ. of Massachusetts,Amherst, MA; E. H. Ruspini, Artificial Intelligence Center, SRI In-ternational, Menlo Mark, CA.1557-6418/06/$17.00c ° 2006 JAIF 1. INTRODUCTION Situation assessment (SA) is an important part of the information fusion (IF) process because it (1) is thepurpose for the use of IF to synthesize the multitudeof information, (2) provides an interface between theuser and the automation, and (3) focuses data collectionand management. Hall and Llinas (Table I) have listeda variety of techniques that need to be solved for SAto be viably implemented in real systems [15]. Sincethe late 1990s there has been few cumulative updates inthe progress of SA and still there are remaining issuesand challenges. During the FUSION05 conference, IvanKadar organized, moderated, and participated in a paneldiscussion with invited leading experts to elicit andsummarize current issues and challenges in SA that needto be researched in the next decade. 1.1. Panel Participants, Topics, and Perspectives This paper serves as a retrospective view of thepanel discussion that was held in July 2005. In thisformat, we list our retrospective and annotated view of the panel information in a condensed (bulletized) formatto make it easier for the reader to assimilate the generalconcepts. Due to space limitation, only a few key issuesare expanded on in text format. ² Organizer: Ivan Kadar, Interlink Systems Sciences,Inc. ² Co-Organizers: Subrata Das, Charles River Analyt-ics and Mieczyslaw M. Kokar, Northeastern Univer-sity ² Moderators: Ivan Kadar, Interlink Systems Sciences,Inc. and James Llinas, SUNY at Buffalo ² July 26, 2005 FUSION 2005–The 8th Interna-tional Conference on Information Fusion, July 25—28,Philadelphia, PA PARTICIPANTS AND PRESENTATION TITLES ² “Knowledge Representation Issues in Perceptual Rea-soning Managed Situation Assessment” Ivan Kadar,Interlink Systems Sciences, Inc., Lake Success, NY ² “Knowledge Representation Requirements for Situ-ation Awareness” John Salerno, Douglas Boulware,Raymond Cardillo, Air Force Research Laboratory,Rome Research Site, NY ² “Situation Assessment: Procedural versus Logical”Mieczyslaw M. Kokar, Department of Elect. & Com-puter Eng., Northeastern University, Boston, MA ² “Tactical Situation Assessment Challenges and Impli-cations for Computational Support” Gerald M. Pow-ell, U.S. Army RDECOM CERDEC I2WD, Ft. Mon-mouth, NJ ² “Situation Assessment in Urban Combat Environ-ments” Subrata Das, Charles River Analytics, Inc.,Cambridge, MA 122 JOURNAL OF ADVANCES IN INFORMATION FUSION VOL. 1, NO. 2 DECEMBER 2006  Report Documentation Page Form Approved OMB No. 0704-0188  Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering andmaintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, ArlingtonVA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if itdoes not display a currently valid OMB control number.   1. REPORT DATE   07 OCT 2006   2. REPORT TYPE   3. DATES COVERED   00-00-2006 to 00-00-2006 4. TITLE AND SUBTITLE   Issues and Challenges in Situation Assessment (Level 2 Fusion)   5a. CONTRACT NUMBER   5b. GRANT NUMBER   5c. PROGRAM ELEMENT NUMBER   6. AUTHOR(S)   5d. PROJECT NUMBER   5e. TASK NUMBER   5f. WORK UNIT NUMBER   7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)   Air Force Research Laboratory,Wright Patterson AFB,OH,45433   8. PERFORMING ORGANIZATIONREPORT NUMBER   9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)   10. SPONSOR/MONITOR’S ACRONYM(S)   11. SPONSOR/MONITOR’S REPORTNUMBER(S)   12. DISTRIBUTION/AVAILABILITY STATEMENT   Approved for public release; distribution unlimited   13. SUPPLEMENTARY NOTES   14. ABSTRACT   15. SUBJECT TERMS   16. SECURITY CLASSIFICATION OF:   17. LIMITATION OFABSTRACT   Same asReport (SAR)   18. NUMBEROF PAGES   18   19a. NAME OFRESPONSIBLE PERSON   a. REPORT   unclassified   b. ABSTRACT   unclassified   c. THIS PAGE   unclassified   Standard Form 298 (Rev. 8-98)  Prescribed by ANSI Std Z39-18  TABLE ISA Challenges and Limitations–Hall and Llinas, [15]JDL Process Processing Description Current Status Challenges and LimitationsLevel 2 Develops a description of currentrelationships among objects andevents in the context of theenvironment (i.e., situationassessment)Numerous prototypesDominance by Knowledge-BasedSystems (KBS)–Blackboard methods–Rule-based representation–Logical templatesKBS experiments–Case based reasoning, Fuzzy LogicNon-real time implementationDominated by prototypesNo experience on scaling to fieldmodels“Excedrin” cognitive modelsDifficult KB developmentPerfunctory Test & EvaluationIntegration of identity/kinematic data ² “Representation and Contribution-Integration Chal-lenges in Collaborative Situation Assessment” DanielD. Corkill, University of Massachusetts, Amherst,MA ² “Human-Aided Multi-Sensor Fusion” Enrique H.Ruspini, et al., Artificial Intelligence Center, SRI In-ternational, Menlo Mark, CA ² “DFIG Level 5 (User Refinement) issues supportingLevel 2 (Situation Assessment)” Erik Blasch, AFRL,WPAFB, OH 1.2. Common Themes While discussion of individual research results bythe participants highlighted specific key issues, therewere common themes that resulted from the panel dis-cussion. The common themes were: COMMON ISSUES ² User–The SA process includes perceptual, interac-tive, and human control ² Process models–updating behavioral models (e.g.Bayes Nets, procedural/logical, perceptual, learning) ² Context–operational situation (i.e., dependent on thecurrent state of the environment) ² Meaning–semantics and syntax issues (formal meth-ods, ontologies) ² Metrics–develop a standard set of metrics (e.g. trust,bounds, uncertainty) COMMON CHALLENGES ² Explanation of process–evidence accumulation andcontradiction in knowledge representation and rea-soning ² Graphical displays to facilitate inferential chains, col-laborative interaction, and knowledge presentation ² Interactive control for corrections and utility assess-ment for knowledge management 2. SITUATION AWARENESS/SITUATION ASSESSMENT There are two main communities that are look-ing at situational information (i.e., Situation Awareness Fig. 1. Endsley’s situation awareness model. (SAW) and Situation Assessment (SA)): the human fac-tors community and the engineering information fusion(IF) research community. SAW is a mental state whileSA supports (e.g. fusion products) that state. The humanfactors notion of SAW is being lead by Mica Endsley[12]. For the IF society, there are many leading peopleproposing different aspects of SAW research. Researchis a way to categorize developments, but another way isby applications. There are many application communi-ties looking at SAW research including: military, med-ical, aviation, security, and environmental. Each mighthave differences, but the commonality rests in the factthat a multitude of data needs to be synthesized into asingle operating picture (dimensionality reduction) [37].Likewise, the salient information needs to be providedto the user to assist the user in completing their missiontasks. 2.1. Situational Awareness Models The Human in the Loop (HIL) of a semi-automatedsystem must be given adequate situation awareness.According to Endsley “SAW is the perception of theelements in the environment within a volume of timeand space, the comprehension of their meaning, and theprojection of their status in the near future.” [12]. Thisnow-classic model, shown in Fig. 1, translates into 3levels: ² Level 1 SAW–Perception of elements in the envi-ronment BLASCH ET AL.: ISSUES AND CHALLENGES IN SITUATION ASSESSMENT (LEVEL 2 FUSION) 123  Fig. 2. Fusion situation awareness model [4]. ² Level 2 SAW–Comprehension of the current situa-tion ² Level 3 SAW–Projection of future statesOperators of dynamic systems use their SAW in de-termining their actions. To optimize decision making,the SAW provided by an IF system should be as preciseas possible as to the objects in the environment (Level 1SAW). A SA approach should present a fused represen-tation of the data (Level 2 SAW) and provide support forthe operator’s projection needs (Level 3 SAW) in orderto facilitate the operator’s goals. From the SA modelpresented in Fig. 1, workload is a key component of themodel that affects not only SAW, but also the decisionand reaction time of the user. 2.2. User Fusion Model As another example, the Situational Model compo-nents [32], shown in Fig. 2, developed by Roy, showthe various information needs to provide the user withan appropriate SAW. To develop the SA model further,we note that the user must be primed for situations tobe able to operate faster, and more effectively.A fusion system must satisfy the user’s functionalneeds and extend their sensory capabilities. Of inter-est to the information fusion community are IF sys-tems which translate data about a region of interest intoknowledge, or at least information over which the hu-man can reason and make decisions. A user fuses dataand information over time and space and acts throughtheir world mental model–whether it be in the heador with graphical displays, tools, and techniques. Thecurrent paradigm for fusion research, shown in Fig. 3,is called the user-fusion model [5]. Fig. 3. User fusion model. 2.3. Perceptual Reasoning Managed Situation Assessment  “Knowledge Representation Issues in PerceptualReasoning Managed Situation Assessment” Ivan KadarThe IF community has had several definitions of SAover time. The JDL Model [14], defined SA as “estima-tion and prediction of relations among entities, to in-clude force structure and cross force relations, commu-nications and perceptual influences, physical context,etc.” DSTO [11, 22] defined SA as “an iterative processof fusing the spatial and temporal relationships betweenentities to group them together and form an abstractedinterpretation of the patterns in the order of battle data.”Issues with the SA definitions, and some subsequentmodels based on these definitions are: ² not domain independent, ² do not incorporate human thought processes, humanperceptual reasoning, the ability to control sensingand essence of response time, 124 JOURNAL OF ADVANCES IN INFORMATION FUSION VOL. 1, NO. 2 DECEMBER 2006  Fig. 4. Perceptual reasoning machine. ² imply use of limited a priori information, ² and only imply potential for new knowledge cap-ture.Therefore, the desired properties of SA are: ² One needs the ability to control Levels 1—4 of DataFusion processes for knowledge capture in SA ² SA is to establish relationships (not necessarily hi-erarchical) and associations among entities, it shouldanticipate with a priori knowledge in order to rapidlygather, assess, interpret and predict what these rela-tionships might be; it should plan, predict, anticipateagain with updated knowledge, adaptively learn, andcontrol the fusion processes for optimum knowledgecapture and decision making ² These features are similar to the characteristics of human perceptual reasoning ² Therefore it is conjectured that the “optimum” SAsystem should emulate human thinking as much aspossibleAs a matter of fact, the godfather of the Internetand knowledge representation, Vannevar Bush [8] inhis famous 1945 essay, “As We May Think” stated,op. cit., “The human mind does not work that wayhierarchically. It operates by association.” Spatial andtemporal associations are key ingredients of PerceptualReasoning Model (PRM).The goal is the perceptual reasoning model which isviewed as a “meta-level information management sys-tem,” as shown in Fig. 4. PRM consists of a feed-back planning/resource control system whose interact-ing elements are: “assess,” “anticipate” and “predict”[16—18]. ² Gather/Assess current, Anticipate future (hypothe-ses), and Predict information requirements and mon-itor intent, ² Plan the allocation of information/sensor/system re-sources and acquisition of data through the controlof a separate distributed multisource sensors/systemsresource manager (SRM), ² Interpret and act on acquired (sensor, spatial andcontextual) data in light of the overall situa-tion by interpreting conflicting/misleading informa-tion.Representative elements and knowledge bases, as-sociated with the assess, anticipate and predict PRMmodules, are categorizable into: (1) functions, witheach function further categorized into (a) knowledge re-quired, (b) knowledge acquisition methods, (c) knowl-edge representation approaches, and (d) implementa-tion techniques. Specific knowledge representation andreasoning (KRR) methods were discussed at the panelhighlighting implementation issues and research chal-lenges. Issues for SA1. Knowledge–a priori and current2. PROCESS–anticipate and gather facts3. User queries instantiation4. Fusion System presents Beliefs5. Need a process model interfaceKRR Challenges for SA1. Adequacy of KRR (logic, ontology, algorithmic,probabilistic), how to quantify/measure?2. Expressiveness of models versus tractability of inference3. Managing Complexity (how to bound problem w/incompleteknowledge)4. Data Information (How to manage heterogeneous anduncertain KSs and detect duplicate or incomplete concepts)5. Presentation of knowledge to different users (what ispragmatic?) 2.4. Syntactic Algorithms and Semantic Synonyms “Knowledge Representation Requirements for Situ-ation Awareness” John Salerno, Doug Boulware, RayCardilloFull Spectrum Dominance (FSD), as defined byJoint Vision 2020, is the ability to be persuasive inpeace, decisive in war and preeminent in any form of conflict. FSD cannot be accomplished without the ca-pability to know what the adversary is currently doingas well as the capacity to correctly anticipate the ad-versary’s future actions. This ability of projection is anelement of Situation Awareness [12, 13]. SA has re-ceived increased attention due to its diverse applicationsin a number of problem domains including: asymmet-ric threat, tactical, cyber, and homeland security [14].Salerno, et al. proposes an architecture that combinesthe Endsley and JDL models (shown in Fig. 5) and hasapplied this model to various strategic, cyber and tacticalapplications [35].Through a display, a user can (1) build a modelby either editing an existing template/model or createa new one; (2) activate/de-activate existing models; or(3) view active models and any evidence that has beenassociated with the model over time. Different political,military, economic, social, infrastructure, and informa-tion models can be accessed and the result published (orsubscribed) to. BLASCH ET AL.: ISSUES AND CHALLENGES IN SITUATION ASSESSMENT (LEVEL 2 FUSION) 125
Similar documents
View more...
Search Related
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks