mastering complexity extracting order from chaos 3rd annual san diego regional security conference...
TRANSCRIPT
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Intelligent Integration
The Next Generation of Intelligent, Automated Systems
Jeffrey Wallace and Barbara Hannibal
3rd Annual San Diego Regional Security Conference
September 14-15, 2010
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
About Us
• Unique Systems Engineering and Development– A distillation of over $2B in government R&D
• Partnered with OSD, JFCOM, “Team Orlando”, the 4 services, SRI, Academia, and others
• Solved world-class/grand challenge problems on F-35, CVN-21, healthcare IT, etc.
2
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Underlying Technology Thrusts
• “Nervous System” – component activation, integration, and interoperability
• Knowledge representation and reasoning– Conceptual graphs– Problem solving module integration
• Complex context representation along with reasoning integration and problem solving– Ontologies and other Semantic Web
technologies3
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Problems• Little to no time for critical decisions
– Imperfect information, must adapt to “dirty” environments
• Interactions happening faster than humans can react– Volume of information, battlespace clutter, and
increasing speed: DE, hypersonics, UxVs
• Current Command and Control (C2) is not real-time and the only source of information
4
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Chinese Anti-Ship Missile
Example: Anti-Ship Missile Defense
Increase Overall System Speed and Effectiveness Simplify Installation and Maintenance
SBX
DDG
DF-21
UAS
CIWS
M-Ship
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Summary• Capture needs through application of novel
techniques• Creation of complex, realistic, and scalable networks
of component inter-relationships• Distribution of autonomous controls and monitors• Implementation of complex webs of cause and effect• Dynamic alteration of the component execution
structure • Adaptation and evolution of the system
8
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
ASW
MIW
SUW
IBGWN
011001111000010101000110101110101010010101011110100001111000111100101110101110
1010101010101010011110010010111001010001010
100011100110011010101010101010011110010010101001010
0011
0101
1101
0101
0010
1010
1111
0100
0011
1100
0111
1001
0111
0101
1100
0101
0100
0110
1011
1010
1010
0101
0101
1110
1000
0111
1000
1111
0010
1110
1011
1000
1010
101
0000
1111
0001
111
0010
1110
1011
100
0101
0
1010
0001
1110
0011
1100
1011
1010
1110
0010
1000
1111
1000
0011
1110
101
0111
1001
0111
0101
1100
0101
0
1010
0001
1110
0011
1100
1011
1010
1110
0010
1000
11
Adaptive Mesh
7
10100001111000111100101110101110001010001111100000111110jfdl;alskedj101
Fast and Flexible
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Code Generation ExampleExample: Turret/Fire Control
8
Slew Elevate
Fire When Slew and Elevate are Complete
Process Firing Commands (and Queuing Them)
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Code Generation ExampleTurret: Fire Command
9
void Turret::fire(){ P_VAR P_BEGIN(2)
// Wait until the turret movement is completed
WAIT_FOR(1, slewComplete, -1); WAIT_FOR(2, elevateComplete, -1);
// Fire the weapon, this would activate the real gun
Fire_M256(); RB_cout << "Flash, Boom, Bang, Echo" << endl; fireComplete = 1; P_END}
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Demonstration Main Points
• Web and Real-time ESB Cooperation– RT ESB handles sustained ~ 1000 messages/sec with
modest hardware and complex systems
• Multiple, simultaneous inter-processor comms employed (shared mem, TCP/IP, RS-232 at 19,200 bps, UDP Broadcast)
• Great performance of complex applications (physics-based 3D visualization and motion prediction)
• Large number of diverse databases and data types employed
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Demo Information Flow
ShadowUAV
SystemComponents
Battlespace Entity
Generator
C4ISR Network Interface (RS-232)
Sensor View 2
Sensor View 1
Web Message Service
C2PC Gateway
C2PC
RS-232 Interface
11
C4ISR NetworkWeb-based/Non-real-time
Battlespace NetworkPredominantly real-time
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
UAV System
AirframeElectro-
MechanicalSystems
SuspensionWheels/Tires
Stability Augmentation
System
CommUplink andDownlink
Powerplant
Pilot Controller
ControlStation
Interface
BIASApplication
Interface
12
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Composability Example: EO/IR Camera
Physics-basedImagerModel
Electro-Mechanical
SystemsModeling
Physics-basedAtmospheric
Modeling
Open SceneGraph (OSG)
*3D Modeling*
OpenProducer
*3D Rendering*
OpenThreads
OSGnVidiaPlugin
CONDORProcessing
Engine
CONDORExternal
ApplicationInterface
IR Camera Simulation
13
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Most “intelligent” computer applications use rule-based expert systems (or some variant) as a means of storing expert knowledge, however these systems always have one or more severe limitations that make them unsuitable for future intelligent software development……..
BRITTLE……….ONCE PROGRAMMEDTHEY ARE EXTREMELY HARD TO
MODIFYCOMPLEX……….HARD (Expensive) TO MAINTAIN
SLOW……….CAN’T APPLY LARGE AMOUNTS OF KNOWLEDGE FOR REAL
TIME APPLICATIONS
RIGID……….CAN’T ADAPT OR SCALE WITH CHANGES IN SITUATION
The Problem
The Conceptual Graph approach was designed to overcome all of these common problems
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
KNOWLEDGE STORAGE USING CONCEPTUAL GRAPHS
New conceptual modeling approaches allow knowledge to be represented in an “Object-Relationship” graphical format - A
surprisingly high level of knowledge complexity can be represented using only three components:
Concepts ActorsRelationships
CAT STAT LOCSIT MAT
A Cat sits on a mat Une Chat assis sur une matte
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Observer Actor
Conceptual Graphs
CONCEPTUAL REASONING ENGINE (CORE)
Stored Knowledge (Reusable)
INPUTS
OUTPUTS
OBSERVER ACTORS SENSE THE STATE OF THE KNOWLEDGE IN THE CONCEPTUAL GRAPHS AND TRANSLATE THAT INTO
THE SPECIFIC OUTPUT FORMAT NEEDED BY THE USER
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Human Factors Architecture Development
Synthetic
Battlespace
Psychological Assessment
Behavior Moderator
Identification
Cognitive Architecture Modification
Human Commander
Human Commander
Psychological Characteristics
Behavior Moderators
Cognitive Architecture Moderators
Modifiable Cognitive
Architecture
Psychological Characteristics:
Traits:Performance UnderPressureConcentration StyleSelf-EsteemBehavioral ControlPhysical Orientation
/
CognitiveFocusDistractibilityFlexiblityInternal External Awareness
Human Factors Assessment Human Factors Modeling Model Validation
Behavior Moderators:
TraitsControlIntroversionExtraversionIntellectual ExpressivenessStatesAnxietyFrustrationObsessiveness
CognitiveAttention capacity / speedWM capacity / speedSkill
Cognitive Architecture Moderators:
StructuralArchitecture topology(Weights on intermodule links)Long-term memory
(Content & structure)
ProcessingModule parameters(Capacity and speed )
Inferencing speed & biases(Cue selection & delays)
orSynthetic
Environment
Psychological Assessment
Tools
(War)game
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Internal Processes Analogy: The Heart Beat
• Atria pump blood to ventricles, which contract
• Nonstop contractions are driven by the heart's electrical system
Internal Process: Synchronous or Asynchronous – Intrinsic CapabilitiesInternal Process: Synchronous or Asynchronous – Intrinsic Capabilities
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
External Processes Analogy: Pacemaker
• External process monitors and interacts with an object (i.e., a pacemaker monitors the heart’s rhythm)
• The electric current makes the heart beat within a certain range
External Process: Synchronous or Asynchronous – Monitor and ControlExternal Process: Synchronous or Asynchronous – Monitor and Control
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Internal Events Analogy: Heart Attack
• Internal occurrence without pre-established time scale
• Certain factors cause the occurrence. Blood flow is restricted, or the nerve system, which controls the heart, malfunctions
Internal Occurrence: Irregular Time Scale – Intrinsic CapabilitiesInternal Occurrence: Irregular Time Scale – Intrinsic Capabilities
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
External Events Analogy: Defibrillation
• External event changes a passive object’s state (i.e., a defibrillator is used for resuscitation)
• External electrical shock is applied to the heart
• Basic representation method
External Occurrence: Irregular Time Scale – Monitor and ControlExternal Occurrence: Irregular Time Scale – Monitor and Control
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Conceptual Graphs
ConceptsConcepts ActorsActorsRelationshipsRelationships
CAT STAT LOCSIT MATCAT STAT LOCSIT MATMAT
A Cat sits on a matA Cat sits on a mat Une Chat assis surune matteUne Chat assis surune matte
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
Basic CG Formation Rules
Concept 1
Relationship 1
Relationship 1
Concept 2
Concepts or Other Actors
Actor 1
Actors or Relationships
Actors or Relationships
Concept 1
Concept
Concept or other Actors
Concept 1
Relationship 1
Actor 1
INVALID
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
CG ExecutionDividend: 9.0
Divisor: 4.0
Number: 144.0
Quotient: 2.0
Remainder: 1.0
SquareRoot: 12.0
Sum: 14.0
divide plus
sqrt
Dividend: 9
Divisor: 4
Number: 144
Quotient: 2
Remainder: 1
SquareRoot: 12
Sum: *s
divide plusplus
sqrt
IF ?r = 0IF ?r = 0
T
TT
T
TT
TT
T
F
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
CG Execution Example
runs
Is a
generating
for
Responsible for
of Producing
AMachine
AnInterpreter
Algorithm
AnInterpretation
An Agent
A Graph A Behavior
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
IBM Sun Oracle BEA RedHat II
Web-based SOA Stack X X X X X X
Heterogeneous Data Source Support X X X X X X
Semantic Web Integration X X X X X
Cloud Integration X X X X
Analog to Digital Interface Support X X
Legacy System Support X
Does Not Require Global Data Model X
Rapidly Adaptable to Change X
Two Day Learning Curve X
Automated Interoperability Assessment X
Sub-Microsecond Latency X
Hard Real-Time Performance X
Support 1M Active System Components X
Capability is Operational X X X X X X
IT Competitor Comparison
Mastering Complexity
Extracting Order from Chaos
3rd Annual San Diego Regional Security Conference
LM NG Boeing GD Raytheon II
Integration Middleware/Dev Env X X X X X X
Heterogeneous Data Source Support X X X X X X
Semantic Web Integration X
Cloud Integration X
Analog to Digital Interface Support X X X X X X
Legacy System Support X
Does Not Require Global Data Model X
Rapidly Adaptable to Change X
Two Day Learning Curve X
Automated Interoperability Assessment X
Sub-Microsecond Latency X
Hard Real-Time Performance X X X X X X
Support 1M Active System Components X
Capability is Operational X X X X X X
LSI Competitor Comparison