1 INTELLIGENCE IN NATUAL AND CONSTRUCTED SYSTEMS
1.1 Introduction,1
1.2 Brief Overview of the Evolving Concepts of Mind and Intelligence,4
1.3 Intelligent Systems:Can We Distinguish Them from Nonintelligent Systems?,12
1.4 Intelligenct:Product and Tool of Behavior and Communication,16
1.4.1 Advantageous Behavior,16
1.4.2 Efficient Symbolic Representation,17
1.4.3 Elementary Loop of Functioning ,19
1.5 Evoluton of Automatisms,20
1.5.1 Learning Artomaton,20
1.5.2 Concept of Automatism and How It Can Be Learned, 20
1.5.3 From Reflexes and Rules to Programs,22
1.5.4 From Programs to Self-Organization,23
1.6 From Agent to Multiscale Communities of Agents,24
1.6.1 The Concept of "Agents"and Its Place in the State of the Art,24
1.7 Cognitive Agents and Architectures,29
References,32
Problems,36
2 THEORETICAL FUNDAMENTALS
2.1 Mathematical Framework of the Architectures for Intelligent Systems,38
2.1.1 Role of Discrete Mathematics in the Development of the Formal Theory of Intelligent Systems,38
2.1.2 What Are the Objects to Which Discrete Mathematics Is Applied?40
2.1.3 What Jobs Does Discrete Mathematics Do?What Are Our Objectives?,41
2.2 Formal Model of Intelligent Systems and Processes, 42
2.2.1 Fundatmental Procedures of Multiresolutional Calculus, 44
2.2.2 Existing Definitions of a Set ,46
2.2.3 What Is a State?How Is Related to the Concept of Object? What Is Change? 50
2.2.4 Formation of New Objects via the GFS Triplet,51
2.2.5 Relations among the Objects :What Is a Relation?What Are the Properties of Relations?Can We Measure Relations?,54
2.2.6 Representation and Reality,54
2.2.7 Models in the Form of Automata:Primitive Agents,56
2.2.8 Multiresolutional Automata,58
2.2.9 Automony and Goal-Orientedness of Intelligent Systems,60
2.3 Necessary Terminology and Assumptions,60
2.3.1 Coordinates,Scope,and Resolution,60
2.3.2 States and State-Space,62
2.3.3 From Objects toward Entites ,64
2.3.4 Clusters and Classes,64
2.3.5 Distinguishability,65
2.3.6 Representation,66
2.4 Construction and Properties of Objects,67
2.4.1 Formation of New Categories,67
2.4.2 Recursive Hierarchies and Heterarchies of Representation,68
2.5 Extracting Entities from Reality ,70
2.5.1 Natural Grouping of Components,70
2.5.2 Grouping within Scientific Processes,73
2.5.3 Grouping Leads to the Multiresolutional Architecture ,76
2.5.4 Differences between Abstraction,Aggregation,and Generalization,79
2.6 Grouping+Filtering+Search:The Elementary Unit of Self-Organzation,80
2.6.1 Concept of GFS,80
2.6.2 Functioning of GFS,81
2.7 Relative Intelligence and Its Evolution,83
2.8 On the Resemblances among Processes of Structuring in Nature and Representation,87
2.8.1 Issue of Resemblance,87
2.8.2 On the Resemblance among the Algorithms Algorithms Applied for Structuring,88
2.9 Semiotic Framework of the Architectures for Intelligent Systems,89
2.9.1 What Is Semiotics?,89
2.9.2 Semiotic Closure ,90
2.9.3 Semiosis:The Process of Learning in Semiotic Systems,92
2.9.4 Reflection and Consciousness,94
References, 96
Problems, 97
3 KNOWLEDGE REPRESENTATION
3.1 Problem of Representing the Natural Worle,100
3.1.1 Unbearable Richness of the Reality,100
3.1.2 Epistemology,102
3.1.3 Evolution of the Theories of Knowledge,103
3.1.4 Semiotics and Future Perspectives,106
3.2 What Is Knowledge?，108
3.2.1 Knowledge as a Phenomenon,108
3.2.2 Knowledge-Related Terminology,109
3.2.3 Storing the Konwledge,112
3.2.4 Why Does the Need for Representing Kownledge Emerge?,113
3.3 Kowledge Representation in the Brain:Acquiring Automatisms,115
3.3.1 An Elementary Information Processing Unit:A Neuron,115
3.3.2 A Ssytem for Seqrching and Storing Patterns: A Neuron,115
3.4 Sensory and Sybolic Representations i the Brain,117
3.4.1 General Comments,117
3.4.2 Sensory Images,117
3.4.3 Symbolic Representations in the Brain,119
3.5 Refernce Frame,Imagination,and Insight,120
3.6 Principles of Konwledge Representation,Entities ,and Relational Structures,121
3.6.1 Nested Hiearchical Konwledge Organization,122
3.6.2 Definitions and Premises of the Principles of Knowledge Representation,124
3.6.3 Definttions of Knowledge-Related Mechanisms of ELF Functioning,133
3.6.4 Types and Classes of Entities, 136
3.6.5 General Characterization of the D-Structure,140
3.6.6 Properties of Labels,143
3.7 Multiresolutional Character of Knowledge and Its Complexity,147
3.7.1 State-Space Decomposition,147
3.7.2 Accuracy,148
3.7.3 Nestedness of Knowledge,149
3.7.4 Recursive Algorithm of Constructing Multiscale Lnowledge Repressentations,150
3.8 Virtual Phenomena of Knowledge Representation,151
3.8.1 Representationfor Immediate Sensory Processing ,151
3.8.2 Intermediate Representation,152
3.8.3 Long-Term Memory Representation,153
Referenes,153
Problems,156
4 REFERENCE ARCHITECTURE
4.1 Components of a Reference Architecture,158
4.1.1 Actuators,158
4.1.2 Sensors,159
4.1.3 Sensory Processing ,159
4.1.4 World Model,15
4.1.5 Value Judgment,159
4.1.6 Behavior Generation,160
4.2 Evolution of the Reference Architecture for Intelligent Systems,160
4.3 Hierarchy with Horizxontal "In Level "Connections,163
4.4 Levels of Resolution,165
4.5 Neural Components of the Architecture,169
4.6 Behavior-Generating Hierarchy,171
4.7 Analysis of Multiresolutional Architectures,172
4.7.1 Elementary Loop of Functioning (ELF),172
4.7.2 Primary Decomposition of an ELF,175
4.7.3 Hierarchies of ELFs:the Essence of NIST-RCS,177
4.7.4 Integrated NIST-RCS Modules,178
4.8 Agent-Basesd Refernce Architectures,180
4.8.1 Elements of Intelligent Software,180
4.8.2 Functioning of the Agent-Based Level,181
References ,184
Problems,186
5 MOTIVATIONS,GOALS,AND VALUE JUDGMENT
5.1 Intermal Needs versus External Goals,188
5.1.1 Neurophysiological Models,188
5.1.2 From Instinct to Motivation to Drive and to Emotion,191
5.1.3 Motivation,192
5.1.4 From Motives to Goals,193
5.1.5 Variors Approaches,194
5.1.6 Development of the Concept of "Goal",196
5.1.7 Cognitive Theories of Goal Formation and Comprison,197
5.2 Value Judgments,199
5.2.1 General Definitions,199
5.2.2 Limbic System,200
5.2.3 Value State-Variables,201
5.2.4 VJ Modules,204
5.2.5 Value State-Variable Map Overlays,207
5.3 Achieving the Goal :Optimization via the Calculus of Variations,207
5.3.1 Notation and Basic Premises,207
5.3.2 A Linearized Third-Oder Plant:Model of a DC Motor,208
5.3.3 Optimization via the Calculus of Variations,209
5.3.4 Results and Discussion,216
References,218
Problems,219
6 SENSORY PROCESSING
6.1 In-level and Inter-level Processes,220
6.1.1 Focusing of Attention(F of GFS),221
6.1.2 Creation of Grouping Hypotheses(G and S of GFS),223
6.1.3 Computation Attributes for Grouping Hypotheses(G of GFS),224
6.1.4 Selection and Confirmation(S of GFS),225
6.1.5 Classification,Recognition,and Organization of Entities,226
6.2 Sensory Processing as a Module of the Level ,228
6.2.1 Information Sources,230
6.2.2 State-Space Tessellation:Sampling,231
6.2.3 Noise,Uncertainty,and Ambiguity,232
6.2.4 Creation of Hypotheses and Testing ,233
6.2.5 Initialization,236
6.3 Hierarchy of Sensory Processing,238
6.3.1 Naturally Emerging Hierarchies of Representation in SP,238
6.3.2 Processing at the Two Adjacent Levels of the Highest Resolution,240
6.3.3 What Happents at the Levels Above?247
6.4 Multiresolutional Nature of Sensory Processing ,251
References,253
Problems,255
7 BEHAVIOR GENERATION
7.1 Preliminary Concepts of Multiresolutional Behavior Generation,257
7.1.1 Definitions,257
7.1.2 Behavior Generation as a Recursive Synthesis of Instantiations from Generalizations,263
7.2 BG Architecture,265
7.2.1 Virtual Loops,265
7.2.2 Real-Time Control and Planning :How They Are Affected by the Sources of Uncertainty,270
7.2.3 Nesting of the Virtual ELFs,271
7.3Srategy of Multiresolutional Control:Generation of a Nested Hierarchy,273
7.3.1 Off-line Dcision-Making Procedures of Planning -Control,273
7.3.2 Nested Hierarchical Information Refinement during On-line Decision Making ,275
7.3.3 Nested Modules,280
7.4 Overall Organization of Behavior Generation,285
7.4.1 Main Concept of COMPUTING bEHAVIOR,285
7.4.2 BG Modules,288
7.4.3 Realistic Exaples of Behavior Generation,291
7.4.4 Generalization upon Realistic Expmples:A Sketch of the Theory,296
7.4.5 Algorithm of Multiresolutional Hierarchical Planning (NIST-TCS Planner),297
7.4.6 BG Module:An Overview,302
7.5 PLANNER,309
7.5.1 Computation Process of Planning ,309
7.5.2 Epistemology and Functions of the PLANNER,310
7.5.3 Planning in Multiresolutional Space versus Planning in Abstraction Spaces ,312
7.5.4 Planning in the Task Space versus Motion Planning,315
7.5.5 Reactive versus Dliberative Decision Making,316
7.5.6 What Is Inside the PLANNER?,317
7.6 EXECUTOR :Its Structure and Functioning,330
7.6.1 Processing the Results of Planning ,330
7.6.2 Structure of EXECUTOR,332
7.6.3 Operations of the EXECUTOR,334
7.6.4 EXECUTOR as a TASK G ENERATOR,336
7.7 Conclusions:Integrating BG in the Intelligent System,337
Referenes,338
Problems,341
8 MULTIRESOLUTIONAL PLANNING:A SKETCH OF THE THEORY
8.1 Introduction to Planning,343
8.1.1 Overview of the Key Results in the Area of Planning,343
8.1.2 Definitions Related to Planning,345
8.1.3 Planning as a Stage of Control,347
8.2 Emerging Problems in Planning, 349
8.2.1 Generic Problems of Behavior Generation,349
8.2.2 Structural Sources of Problems,350
8.2.3 Representaion and Planning,350
8.2.4 Classificaion of Planning Problems in Intelligent Systems ,352
8.3 Planning of Actions and Planning of States,354
8.3.1 Algorithms of Planning,354
8.3.2 Visibility-Based Planning ,354
8.3.3 Local Planning :Potential Field for World Representation,356
8.3.4 Golbal Planning:Search for the Trajectories,356
8.4 Linkage between Planning and Learning,357
8.4.1 Learing as a Source of Representation,357
8.4.2 Interrelations between Planning and Learning,358
8.5 Planning in Architectures of Behavior Generation,358
8.5.1 Hierarchical Multiresolutional Organization of Planning,358
8.5.2 Case Study:A Pilo for an Autonomous Vehicle,360
8.6 Path Planning in a Multidimensional Space,36
8.6.1 State-Space Representation,366
8.6.2 Expert Rules/Heuristics,368
8.6.3 Techniques to Recduce Computation Time, 370
8.6.4 Experimental Results, 371
8.7 Multiresolutional Planning as a Tool of Increasing Efficiency of Behavior Generation,373
8.7.1 Multiresloutional Planning Embodies the Intelligence of a System,373
8.7.2 Multiresolutional Planning Reduces the Complexity of Computations,374
8.7.3 Applying the General S3-Search Algorithm for Plannint with Complexity Reduction,379
8.7.4 Tessellation,381
8.7.5 Testing of the Representation,390
8.7.6 Search,387
8.7.7 Consecutive Refinement,390
8.7.8 Evaluation of Complexity,393
References,395
Problems,398
9 MULTIRESOLUTIONAL HIERARCHY OF PLANNER/EXECUTOR MODULES
9.1 Hybrid Control Heerarchy,399
9.2 Theoretical Premises,401
9.3 Canonical Hybrid PLANNER/EXECUTOR Module,404
9.3.1 Basic Control Architecture,404
9.3.2 Recursive Application and Nestion of the Canonical Hybrid PLANNER/EXECUTOR Modules,406
9.3.3 World Model :Maintenance of Knowledge,407
9.3.4 Nesting and Bandwidth Separation,410
9.4 Quasi-minimum Time Functioning of PLANNER/EXECUTOR Modules Equipped with a Production System,413
9.4.1 Assumptions about System Dynamics,413
9.4.2 Assumptions about Propulsion Force DEVELOPMENT ,416
9.4.3 Assumptions about the Resistance Force,417
9.4.4 Deterining Standard Components of the Control Cycle,418
9.4.5 Production System for Combining the Control Cycle from Standard Components,420
9.4.6 Simulation of a Real System,422
9.5 Approzimate Inverse via Forward Searching ,427
9.5.1 Introduction,427
9.5.2 Concept of Approximate Imnverse Image of the Reference Trajectory,428
9.5.3 Approximate Inversion via Search Procedures,430
9.5.4 Experimental Results,433
9.6 Multirate Hierarchical Predictive Control System,439
9.6.1 Introduction,440
9.6.2 Hierarchical Structure with Averaging the Wavelet Algorithm,441
9.6.3 Hierarchical Controller Design,443
9.6.4 Simulation Reults,444
9.6.5 Multirate Hierarchical Controller,447
9.7 Conclusion,452
References,453
Problems,455
10 LEARNING
10.1 Intelligent Systems and Learning ,457
10.2 Definitions of Learning,458
10.3 Implicit and Explicit Logical and Psychological Schemes of Learning ,463
10.3.1 Need in"Bootstrap"Knowledge:Azxioms and "Self-evident" Principles,463
10.3.2 Gestalt Principles:Entity Discovering Insights,464
10.3.3 Repetitiveness:Induction,and Deduction,467
10.3.4 Recursion and Iteration,467
10.3.5 Typology of Learning,468
10.3.6 On the Difference between Adaptive and Learning Systems,477
10.4 Information Acquistition via Learning :Domains of Application,477
10.4.1 Estimation and Recognition as Components of Learning,477
10.4.2 Domains of Application for the Theory,479
10.5 Axiomatic Theory of Learning Control Systems ,481
10.5.1 Axioms,481
10.5.2 Learning Control Systems,487
10.5.3 Emerging Nonconventional Issues,490
10.5.4 Exosystem and Multi-actuation,493
10.6 Leaning and Behavior Generation:Constructing and Using MR Representation and Goal Hierardhies,494
10.6.1 Learning from Multiple Experiences ,494
10.6.2 Algorithms of Unsupervised Learning,496
10.6.3 Evolution of Knowledge,507
10.6.4 Focusing of Attention,508
10.6.5 Formation of Similarity Clusters,509
10.6.6 Searching for Valid Hypotheses among Clusters,510
10.6.7 Learning in Behavior Generation,511
10.6.8 Evoluton fo ultiresolutional Learning Automata,513
10.6.9 Further Research in LPA,515
10.7 Baby-Robot:Analysis of Early Cognitive Development,517
10.7.1 ECD:Its Significance for Learning in ntelligent Systms,517
10.7.2 Resemblances and Differences between ECD in Human and ISs,518
10.7.3 Quantitative Learning Domain,520
10.7.4 Conceptual Learning Domain,520
10.7.5 Baby-Robot:Simulation Tool for Analysis of ECD Processes,523
10.7.6 Baby-Robot:A Mental Experiment,524
10.7.7 A Possible Algorithm of Learning,525
10.7.8 Conditions of Clause Generation,527
10.7.9 Learning When the Goal Is Given,528
10.7.10 Simulation and Physical Experiments with Baby-Robot,528
10.7.11 Implications of Baby-Robot Research,535
10.8 Applying Neural Networks:Architectures for Generalization of Multiple Information,536
10.8.1 Neural Networks:Architectures for Generalization of Multiple Information,536
10.8.2 CMAC :An Associative Neural Network Alternative to Backpropagation,540
References,549
Problems,559
11 APPLICATIONS OF MULTIRESOLUTIONAL ARCHITECTURES FOR INTELLIGENT SYSTEMS
11.1 An Intelligent Systems Architecture for Manufacturing ,560
11.1.1 ISAM Standard,560
11.1.2 ISAM as a Conceptua Framework,561
11.1.3 ISAM versus Current Practice,561
11.1.4 ISAM as a Reference Model Architecture,562
11.1.5 AnISAM Example,563
11.2 Planning in the Hierarchy of NIST-RCS for Manrfacturing,568
11.2.1 Planner within a Sinle RCS Module,569
11.2.2 Planner in the RCS Hierarchy,570
11.2.3 Algorithm of Planning,571
11.2.4 RCS Approach within the System of Contemporary View on Scheduling in Manufacturing,573
11.3 Inspection Workstation-Based Testbed Application for the Intelligent Systes Architecture for Manufacturing ,575
11.3.1 HPCC Program:A Paradigm for ISAM Implementation,576
11.3.2 The Approach,576
11.3.3 Applicaltion Design for Prototype System Implementations,577
11.3.4 Modules,578
11.4 RCS-Based RoboCrane Integration,585
11.4.1 The Prototypes,586
11.4.2 RoboCrane Subsystems,588
11.5 Mobile Robot Goes Multiresolutional,592
11.5.1 Types of Decision Making ,596
11.5.2 Rule Bases,596
11.5.3 Precies Preplanning (PP)Algorithm,597s
11.5.4 Implementation.600
11.5.5 Instantaneous Decision Maker(ID)and Path Monitor,600
11.6 Mission Structrre for an Unmanned Vehicle,601
11.6.1 Levels of the System,602
11.6.2 Classification of Missions,606
11.6.3 Activities of the Platoon Leader,609
11.6.4 Tasks Assigned to a Section,611
11.6.5 Tasks Assigned to a Single Vehicle,612
11.6.6 Tasks Assigned to A Vehicle's Subsystems,612
11.6.7 Invariant Elements of Mission Programs,613
11.6.8 Detection and Recognition of Sensed Objects,615
11.7 4-D/RCS as an Implementantion Guide for Demo III,615
11.7.1 Subset of 4-D/RCS for Demo III,615
11.7.2 Examples of 4-D/RCS Entities,628
11.7.3 Integration of the System,635
References,639
12 INTELLIGENT SYSTEMS:PRECURSOR OF THE NEW PARADIGM IN SCIENCE AND NEGINEERING
12.1 Multidisciplinarity:The Most Promising But a Bumpy Road,642
12.2 What Is the Core Activity of Intelligent Systems?,647
12.3 More Questions Than We Can Handle,649
12.3.1 Logic and Automata:Reasoning about Truth and Equivalence,649
12.3.2 Models That Incorporate the Concept of Goal,651
12.3.3 An Ability to Do Unprescribed Actions,652
12.4 Multiresolutional Recursive Sign Processing with Learning (MRSPL Processing ),652
12.4.1 What Are the Tools?,652
12.4.2 Semiotics and Objce-Oriented Engineering :Kindred Strangers,653
12.4.3 Unit of Intelligence,654
12.4.4 Unit of Life?,654
12.4.5 How to Use It,655
12.4.6 Invariance,656
12.5 Multiresolutional Intelligence,657
12.6 Learning How to Know:Semiotics and Multiscale Cybernetics,659
12.7Engineering of Mind,662
References,665
ABBTREVIATIONS
NAME INDEX
INDEX
鄂公網(wǎng)安備 42010302001612號(hào) 