9th International Command and Control Research and Technology Symposium
Multi-Hypothesis Structures and Taxonomies for Combat Identification Fusion Tod M. Schuck, J. Bockett Hunter, Daniel D. Wilson September 2004 Copenhagen Denmark Paper #141
Agenda Presentation Objective Architecture Definition for Disparate Information Taxonomy f-refinement Partition Refinement Canonical Mapping Response Mapping CID Realization of JDL Model CID Bayesian Network CID Situational Awareness Fusion Expansion Conclusions and Future Work Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -2
Presentation Objective We want to answer the question, “How can we establish the relationship between information sets such as needed for a fusion process????” Address movement of information across multiple hypothesis classes
Relate it to developing the identification of objects Describe how it can be combined both within and
between JDL levels
Result will be an information architecture that is naturally adaptive to information regardless of quality, level, specificity
Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -3
An Environment of Disparate Information…
Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -4
Solution Requires Good Architecture Definition Defining architecture requires investigation into detailed taxonomic relationships between information sets and subsequent canonical mappings
Subsequent response mapping can be defined Results can be tied into JDL model Taxonomy – a classification scheme for objects with
mutually exclusive labels (parallels study of ontologies) CID {Friend, Assumed Friend, Hostile, etc.} Nationality {US, Russia, France, Iraq, etc.} Category {Air, Sea, Land, etc.} Platform {Fighter, Bomber, Civil, etc.} Type {F-14, F/A-18, F-22, etc.} Class {F-14A, F/A-18D, etc.}
Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -5
Taxonomy f-refinement Taxonomy A is an f-refinement of taxonomy B if: : A → B such thatb1if≠ b2 f −1 (b1 ) ∩ f −1 (b2 ) = ϕ where ϕ = empty set
f is a functionf
A
then
F-16A F-16B F-16C F-16D F/A-18A F/A -18B F/A -18C F/A -18D F/A –18E F/A –18F f
B
F-16
F/A-18
If a taxonomy of A is an f-refinement of taxonomy B and a ∈ A, b ∈ B, and f(a) = b, we say that a is an f-refinement of b Example: F/A-18A in the Class taxonomy is a refinement of F/A-18 in the Type taxonomy
Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -6
Partition Refinement Given a set S of objects, a
taxonomy imposes a partition on the set
Each element of the partition is the set of all elements of S for which a single element of the taxonomy is the appropriate name
Example: an element of the
partition imposed on aircraft by the Type taxonomy is the set of all F-15s, all 747s, etc.
F/A-18E F/A-18F F/A-18A
F-16A
-18 F/A
F-16B F-16D
747
F/A-18D
16 F-
Mig-29
F/A-18C
F-16C F/A-18B
A340
A taxonomy T1 is a refinement of another taxonomy T2 if the partition imposed by T1 is a refinement of the partition imposed by T2
Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -7
Canonical Mapping Canonical mappings provide a way to exploit
information about an object from different taxonomies to categorize the object in one of those taxonomies, or in another, completely different, taxonomy
Set of canonical mappings must be defined between any two related taxonomies
In the case of a collection of
taxonomies that are successive refinements, the canonical mappings reflect the hierarchical nature of the taxonomies themselves Example: sets T6 and T3 are related through both the mapping m3 and the composite mapping m2*m1 (so not a true canonical mapping) Copyright Lockheed Martin Maritime Systems and Sensors
T1
T2
m1 m3
T6 T5
m2 T3
T4
9th ICCRTS Sept 2004 paper #141 -8
Response Mapping Response mapping is a way to interpret a response with
elements from one taxonomy in terms of another taxonomy Provides a means of interpreting a response with elements from more than one taxonomy in the various referenced taxonomies
Example: let R be a response
from a source of information composed of a set of attributes, let the canonical mapping from taxonomy Ti to taxonomy Tj be mij - Each taxonomy potentially has elements that are part of the response (R1 and R2 in the figure), as well as elements that are the images, under a canonical mapping, of elements in other taxonomies (m12(R1), m21(R2), m13(R1), m23(R2)
m13
m12
T1
m21
T2 m12(R1)
m21 12(R21) R1
Copyright Lockheed Martin Maritime Systems and Sensors
R
R2
m23
T3 m13(R1) m23(R2)
9th ICCRTS Sept 2004 paper #141 -9
JDL Model and CID Realization
JDL Model
CID Implementation
Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -10
A priori CID Bayesian Network (Level 1) N a t io n a lit y US 6 7 .6 B e lg iu m 0 D e n m a rk 0 E gypt 0 In d o n e s ia 0 .2 5 Is r a e l 3 0 .9 N e th e r la n d s 0 N o rw a y 0 P a k is ta n 0 P o r tu g a l 0 S in g a p o r e 0 T a iw a n 0 T h a ila n d 0 V e n e z u e la 0 B a h r a in 0 G re e c e 0 K o re a 0 T u rk e y 0 U n ite d A r a b... 0 Canada 0 A u s tr a lia 0 K u w a it 0 F in la n d 0 S w itz e r la n d 0 M a la y s ia 0 Ir a n 0 A r g e n tin a 0 C h in a 0 E n g la n d 0 Japan 0 P o la n d 0 G e rm a n y 0 P h illip in e s 0 M o ro c c o 0 S a u d iA r a b ia 0 Sweden 0 A fr ic a 0 B r a z il 0 F ra n c e 0 S p a in 1 .2 6
Given a-priori state
of universe for F-14, F-16, F/A-18 and Boeing 737 aircraft
ID 8 .7 8 F14A 0 .5 7 F14B 0 .5 7 F14D 2 0 .0 F16A 3 .4 0 F16B 2 9 .5 F16C 1 5 .3 F16D 1 2 .9 FA18AC 4 .0 0 FA18BD 0 .4 3 FA18E 0 .4 3 FA18F B o e in g 7 3 7 2 0 0 0 .3 2 B o e in g 7 3 7 3 0 0 0 .5 0 B o e in g 7 3 7 4 0 0 0 .6 5 B o e in g 7 3 7 5 0 0 1 .1 4 B o e in g 7 3 7 7 0 0 0 .5 0 B o e in g 7 3 7 8 0 0 0 .4 2 B o e in g 7 3 7 9 0 0 0 .5 6
A ir
C a te g o r y 100 F14 F16 FA18 B o e in g 7 3 7
T yp e 9 .9 2 6 8 .2 1 7 .8 4 .1 0
P la t f o r m F ig h te r 9 5 .9 C o m m e rc ia l 4 .1 0
C la s s 8 .7 8 F14A 0 .5 7 F14B 0 .5 7 F14D 2 0 .0 F16A 3 .4 0 F16B 2 9 .5 F16C 1 5 .3 F16D 1 2 .9 FA18AC 4 .0 0 FA18BD 0 .4 3 FA18E 0 .4 3 FA18F B o e in g 7 3 7 2 0 0 0 .3 2 B o e in g 7 3 7 3 0 0 0 .5 0 B o e in g 7 3 7 4 0 0 0 .6 5 B o e in g 7 3 7 5 0 0 1 .1 4 B o e in g 7 3 7 7 0 0 0 .5 0 B o e in g 7 3 7 8 0 0 0 .4 2 B o e in g 7 3 7 9 0 0 0 .5 6
Shaded areas (Nationality and Platform)
represent where new info will be inputted Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -11
A priori CID Bayesian Network (Level 1) BN after new sensor inputs: Fighter COM N a t io n a l it y US 6 1 .4 B e lg iu m 2 .0 9 D e n m a rk 0 .8 9 E gypt 2 .6 5 I n d o n e s ia 0 .3 3 Is ra e l 3 .2 0 N e th e r la n d s 3 .2 4 N o rw a y 0 .8 9 P a k is ta n 0 .9 4 P o rtu g a l 0 .2 1 S in g a p o r e 0 .6 8 T a iw a n 1 .7 8 T h a ila n d 0 .7 4 V e n e z u e la 0 .2 5 B a h r a in 0 .2 6 G re e c e 1 .8 2 K o re a 2 .2 6 T u rk e y 3 .3 4 U n it e d A r a b. .. 1 .0 1 Canada 1 .6 3 A u s t r a lia 1 .5 1 K u w a it 0 .4 9 F in la n d 0 .7 0 S w it z e r la n d 0 .3 7 M a la y s ia 0 .6 0 Ira n 0 .8 4 A r g e n t in a 0 .1 8 C h in a 0 .4 3 E n g la n d 0 .4 3 Japan .0 6 5 P o la n d 0 .2 1 G e rm a n y 0 .8 3 P h illip in e s .0 5 6 M o ro c c o 0 .3 7 S a u d iA r a b ia 0 .1 6 Sweden 1 .0 4 A fr ic a 0 .1 7 B r a z il 0 .6 2 F ra n c e 0 .3 7 S p a in 0 .9 0
A ir
C a te g o ry 100 F14 F16 FA18 B o e in g 7 3 7 P la t f o r m F ig h te r 7 5 .6 C o m m e rc ia l 2 4 .4
Air (0.15)
Israel
ID 7 .1 1 F14A 0 .4 1 F14B 0 .4 1 F14D 1 7 .3 F16A 4 .0 7 F16B 2 3 .4 F16C 6 .1 0 F16D 1 2 .2 FA18AC 4 .0 7 FA18BD 0 .3 0 FA18E 0 .3 0 FA18F B o e in g 7 3 7 2 0 0 2 . 0 3 B o e in g 7 3 7 3 0 0 2 . 0 3 B o e in g 7 3 7 4 0 0 3 . 0 5 B o e in g 7 3 7 5 0 0 9 . 1 5 B o e in g 7 3 7 7 0 0 2 . 0 3 B o e in g 7 3 7 8 0 0 3 . 0 5 B o e in g 7 3 7 9 0 0 3 . 0 5
US
(0.85)
(0.7)
(0.1)
Indonesia Spain
T yp e 7 .9 3 5 0 .8 1 6 .9 2 4 .4
(0.1)
(0.1)
C la s s 7 .1 1 F14A 0 .4 1 F14B 0 .4 1 F14D 1 7 .3 F16A 4 .0 7 F16B 2 3 .4 F16C 6 .1 0 F16D 1 2 .2 FA18AC 4 .0 7 FA18BD 0 .3 0 FA18E 0 .3 0 FA18F B o e in g 7 3 7 2 0 0 2 . 0 3 B o e in g 7 3 7 3 0 0 2 . 0 3 B o e in g 7 3 7 4 0 0 3 . 0 5 B o e in g 7 3 7 5 0 0 9 . 1 5 B o e in g 7 3 7 7 0 0 2 . 0 3 B o e in g 7 3 7 8 0 0 3 . 0 5 B o e in g 7 3 7 9 0 0 3 . 0 5
Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -12
CID Situational Awareness (SA) Fusion Expansion Level 1 SA: Perception of the environmental elements – The identification of key elements of “events” that, in combination, serve to define the situation
JDL – numeric processing of tactical components SA – symbolic processing of these entities
Level 2 SA: Comprehension of the current situation – This combines level 1 events into a comprehensive holistic pattern (or tactical situation) JDL and SA virtually identical
Level 3 SA: Projection of future status – Projection of the
current situation into the future, so as to predict the course of an evolving tactical situation SA more general than JDL, includes projection of
ownship/aircraft/etc., and friendly intent
Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -13
Tactical Elements Employed by Fusion Level
Ev en ts specific
y tactical elements
aggregated
organizations
En ob em jec y d tiv oct es rin & ca e, pa bil it
options
Who?
Level 2: Situation Refinement
global
global
Be ha vio ra lO bj ec Th ts re a Ta t E cti nvi cs, ro & nm Th en re t, E at Be nem ha y vio rs
Why?
Fr an iend d m ly iss vul ion ne s rab i
Level 3: Threat Refinement
needs
liti
es
What if?
local
local
When?
Level 1: Object Refinement
Ph ys ica lO bj ec ts
What?
individual
Where? Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -14
Proposed Threat Evaluation Tool CID Mission Planning (IPB)
Object Behaviors Track Fusion JDL Level 1 Hypotheses
“Trigger” Thresholds Defensive “Impact” missions, resources, & assets Anticipated scenarios, threats and environments
Capability & effectiveness of class Design mission of class
Source of class
CID Fusion JDL Level 2 Hypotheses
Goals of ID Source of ID
Flight plans, FAA updates, ADS-B
Assessment (Threat Evaluation)
Initial Hypothesis Generation
Assessment/ Comprehension Hypothesis Generation/ Refinement
“Event” Recognition “Object” Aggregation
Knowledge Base
Impact Proj.
Impact Modeling
Hypothesis Manager
Threat Assessment state, confidence, rationale, & source
Alert Generation
I&W Indicators & Warnings
Hypotheses on “Intent” (mission, activity, etc.)
“Mission” Recognition Initial Hypothesis Generation
Threat Assess.
Domain of Level 3 Fusion
Hypothesis Generation/ Refinement
Assessment/ Comprehension
Predictive Profiling Courses of Action
Intent Determination state, confidence, rationale, & source
“Intent” Determination (Mission, Activity) Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -15
Conclusions and Future Work Contextual relationship of information is paramount Fusion process must incorporate these relationships CID information wrt context, time, timeliness, quantity, and quality must be known
Future work: Metrics for information value, completeness, and costs of
decisions can be developed and integrated
Contextual reasoning leading to predictive SA
Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -16
Author Information
Tod Schuck is a Lead Member of the Engineering Staff at Lockheed Martin MS2, Moorestown, NJ. He is the recipient of numerous awards for his technical work including the ‘01 Lockheed Martin MS2 Author of the Year, finalist for best paper by OASD C3I at the 7th International Command and Control Research Technology Symposium (ICCRTS) ‘02, best paper (2nd place) IEEE National Aerospace and Electronics Conference (NAECON) ‘00, and a group award winner of Vice President Al Gore’s “Silver Hammer” Award for Extraordinary Effort for Changing the Way That Government Does Business (July ‘97) as the technical lead for the Mk 17 NATO SeaSparrow radar SDP. His areas of expertise are in identification sensor design and development, fusion algorithm development, information architecture design; and requirements generation, and testing. He holds a BSEE from Georgia Tech, an MSEE from Florida Tech, and is pursuing a Ph.D. from Stevens Institute of Technology in Systems Engineering.
J. Bockett Hunter is a Senior Member of the Engineering Staff at Lockheed-Martin MS2, Moorestown, NJ. He has been on the faculty of Indiana University, worked on Navy operations research problems, and a wide variety of intelligence systems. His areas of expertise include algorithm development, systems engineering for very large systems including system concept definition, requirements development, design, and testing. He holds a BS in mathematics from Caltech and an MA in mathematics from Indiana University.
[email protected]
Daniel D. Wilson is a Principal Systems Engineer at Lockheed-Martin MS2, Burlington, MA. He has worked on Military Operations Research problems in support of numerous DoD programs. His areas of expertise include C4ISR systems, operations analysis, knowledge fusion, predictive situational awareness, dynamic replanning strategy development, collaborative mission planning, system effectiveness modeling, decision analysis, risk analysis, and cost estimation. He holds a BS in Computational Mathematics from the University of Vermont and an MS in Operations Research from Stanford University.
[email protected]
[email protected]
Copyright Lockheed Martin Maritime Systems and Sensors
9th ICCRTS Sept 2004 paper #141 -17
