An ‘ Intelligent Consultant ’ – The Support for C 2 System Operator

The purpose of this paper is to present a concept of so called ‘intelligent consultant’ – that is a tool that provides C&C system operator a support in case the gathered information (acquired from observation means) is incomplete, imprecise or even conflicted. It is the intention of the authors to focus on the methodological aspect of the problem which may be formed as: Does Dezert-Smarandache Theory (DSmT) of plausible and paradoxical reasoning enable to deal with difficult target classification which normally happens in maritime C&C systems. For this reason a comparison of hard-decision fusion, DSmT fusion and a combination of DSmT and ontology fusion algorithms has been established. Some numerical experiments have been made and their results have been delivered in this paper. 1.0 INTRODUCTION Nowadays, one of the most important requirements imposed on maritime C&C systems is a dealing with a problem of integrating information gathered from diverse sources (e.g. radars, video cameras and visual sightings). The diversity of sensors utilised for acquiring information, useful from the tactical point of view, implies a necessity of reconstruction of the situational awareness at the presence of concise information as well as at the presence of uncertain, incomplete or even conflicted information. Experience in working with C&C systems shows that there are particular situations, where an indication of incoherent information (ambiguity or conflict) is caused by imprecise processing of information related to target attributes (e.g. target misclassification or attribute conversion errors). In these particular cases the most often step of the fusion algorithms is holding the combination process and an indication of the existing problem. If the number of targets grows, the probability of necessary manual intervention also rises, which in consequence may lead to a degradation of quality of the situational awareness. An analysis of C&C systems operators’ needs shows that an automatic combination of attribute information is not enough requirement imposed on modern fusion systems. The fact the operator is responsible for the quality of elaborated information does not contradict with the idea of supporting 1 Scientific work, financed with science means from 2007 to 2010 as an ordered research project. 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 and maintaining 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, Arlington VA 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 it does not display a currently valid OMB control number. 1. REPORT DATE OCT 2009 2. REPORT TYPE N/A 3. DATES COVERED 4. TITLE AND SUBTITLE An Intelligent Consultant The Support for C2 System Operator When Fusing Ambiguous Information1 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) R&D Marine Technology Centre Dickmana 62 81-109 Gdynia, Poland 8. PERFORMING ORGANIZATION REPORT NUMBER 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S) 11. SPONSOR/MONITOR’S REPORT NUMBER(S) 12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release, distribution unlimited 13. SUPPLEMENTARY NOTES See also ADB381582. RTO-MP-IST-087 Information Management Exploitation (Gestion et exploitation des informations). Proceedings of RTO Information Systems Technology Panel (IST) Symposium held in Stockholm, Sweden on 19-20 October 2009., The original document contains color images. 14. ABSTRACT The purpose of this paper is to present a concept of so called intelligent consultant that is a tool that provides C&C system operator a support in case the gathered information (acquired from observation means) is incomplete, imprecise or even conflicted. It is the intention of the authors to focus on the methodological aspect of the problem which may be formed as: Does Dezert-Smarandache Theory (DSmT) of plausible and paradoxical reasoning enable to deal with difficult target classification which normally happens in maritime C&C systems. For this reason a comparison of hard-decision fusion, DSmT fusion and a combination of DSmT and ontology fusion algorithms has been established. Some numerical experiments have been made and their results have been delivered in this paper. 15. SUBJECT TERMS 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT SAR 18. NUMBER OF PAGES 14 19a. NAME OF RESPONSIBLE PERSON a. REPORT unclassified b. ABSTRACT unclassified c. THIS PAGE unclassified Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18 An ‘Intelligent Consultant’ – The Support for C2 System Operator When Fusing Ambiguous Information 19 2 RTO-MP-IST-087 UNCLASSIFIED/UNLIMITED UNCLASSIFIED/UNLIMITED him/her with the ‘intelligent consultant’ software which is to present the optimal solution, according to gathered imprecise and incomplete evidences. The purpose of this paper is to present a concept and results of numerical experiments, related to the above mentioned intelligent consultant. This tool effectively utilises the relations among target attributes, defined in sensor network ontology [1]. A precise definitions of these relations have been given using elements of Dezert-Smarandache Theory [2], [3]. 2.0 COMBINATION OF ONTOLOGIES AND DSMT This section presents a proposition of an ontology framework for a sensor network, dedicated to monitor the target threat. In the solution there were utilised concepts and concept lexicons of JC3 model [4]. The authors’ intention was to show the way relations of three attributes (threat, platform and activity) should be defined, rather than to present the complete SN ontology. Table 1 presents a bijective assignment of concepts to elements of a concept lexicon. As it was mentioned before, this assignment need not be a bijection, however it is desirable especially if sets of values for attributes of platform and activity are numerous. Table 1: SN ontology: concepts and concept lexicon. Concepts Concept lexicon Threat An OBJECT-ITEM that is assumed to be a friend because of its characteristics, behaviour or origin. object-itemhostilitystatus-code ASSUMED FRIEND An OBJECT-ITEM that is positively identified as enemy. HOSTILE ...according to JC3 ... according to JC3 Platform General designator for aircraft/multi-role aircraft carrier; surfacevessel-typecategory-code AIRCRAFT CARRIER, GENERAL Craft 40 meters or less employed to transport sick/wounded and/or medical personnel. AMBULANCE BOAT ... according to JC3 ... according to JC3 Activity To fly over an area, monitor and, where necessary, destroy hostile aircraft, as well as protect friendly shipping in the vicinity of the objective area. action-taskactivity-code PATROL, MARITIME Emplacement or deployment of one or more mines. MINE-LAYING ... according to JC3 ... according to JC3 The assignment of relations among attributes to relation lexicons (Table 2) is a surjection. In order to define the relations among attributes DSmT combining and conditioning rules have been applied. The preferred rule for conditioning is the rule no. 12. When combining evidence, there is a possibility to use many combination rules, depending the particular relation. However, for simplicity, it is suggested to apply the classic rule of combination (DSmC), which has properties of commutativity and associativity. An ‘Intelligent Consultant’ – The Support for C2 System Operator When Fusing Ambiguous Information RTO-MP-IST-087 19 3 UNCLASSIFIED/UNLIMITED UNCLASSIFIED/UNLIMITED Table 2: SN ontology: relations and relation lexicon. Relations Remarks Relation lexicon Rel. I: cond(.) Based on DSmT Conditioning  According to distinctive features Implication Rel. II: cond(.) Based on DSmT Conditioning  Based on DSmT Combination Rel. III: cond(.) Based on DSmT Conditioning  Based on DSmT (combination rule need not be identical with one in Relations II) Combination Below, there have been presented examples of particular types of relations. In case of the relation of type I it is possible to reason about a value of a certain attribute, based on the knowledge about the other ones. However, if the unambiguous deduction of the third attribute is not possible, due to the majority of possible solutions, an application of abductive reasoning (selection of the optimal variant) seems to be justified. Relations I: (Threat, Platform)  Activity: (FAKER, FRIGATE TRAINING) TRAIN OPERATIONS; (Threat, Activity)  Platform: (FAKER, TRAIN OPERATIONS) TRAINING CRAFT; (Platform, Activity)  Threat: (HOUSEBOAT, PROVIDE CAMPS) NEUTRAL;