Developing autonomous agent systems : a practical guide to designing, building, implementing and testing agent systems

Developing Intelligent Agent Systems......Page 3
Contents......Page 7
Foreword from the Series Editor......Page 11
Preface......Page 13
Acknowledgement......Page 15
1.1 What is an Intelligent Agent?......Page 17
1.2 Why are Agents Useful?......Page 20
2.1 Situated Agents: Actions and Percepts......Page 23
2.2 Proactive and Reactive Agents: Goals and Events......Page 24
2.3 Challenging Agent Environments: Plans and Beliefs......Page 26
2.4 Social Agents......Page 28
2.5 Agent Execution Cycle......Page 29
2.5.1 Choice of Plan to Execute......Page 30
2.5.2 Many Ways to Achieve a Goal......Page 32
2.6 Summary......Page 35
3 Overview of the Prometheus Methodology......Page 37
3.1 Why a New Methodology?......Page 38
3.2 Prometheus: A Brief Overview......Page 39
3.2.1 System Specification......Page 40
3.2.2 Architectural Design......Page 41
3.2.3 Detailed Design......Page 42
3.3 Guidelines for Using Prometheus......Page 43
3.4 Agent-Oriented Methodologies......Page 45
4 System Specification......Page 49
4.1 Goal Specification......Page 50
4.1.1 Identify Initial Goals......Page 51
4.1.2 Goal Refinement......Page 52
4.2 Functionalities......Page 57
4.3 Scenario Development......Page 59
4.3.1 Goal Step Details......Page 61
4.3.2 Capturing Alternative Scenarios......Page 62
4.4 Interface Description......Page 63
4.4.1 Percepts and Actions......Page 64
4.4.2 Data......Page 66
4.5 Checking for Completeness and Consistency......Page 67
5 Architectural Design: Specifying the Agent Types......Page 69
5.2 Grouping Functionalities......Page 72
5.3 Review Agent Coupling – Acquaintance Diagrams......Page 79
5.4 Develop Agent Descriptors......Page 81
6 Architectural Design: Specifying the Interactions......Page 83
6.1 Interaction Diagrams from Scenarios......Page 84
6.2 Interaction Protocols from Interaction Diagrams......Page 90
6.3 Develop Protocol and Message Descriptors......Page 93
7 Finalizing the Architectural Design......Page 97
7.2 Identifying Boundaries of the Agent System......Page 98
7.3 Describing Percepts and Actions......Page 100
7.4 Defining Shared Data Objects......Page 104
7.5 System Overview Diagram......Page 107
7.6 Checking for Completeness and Consistency......Page 110
7.6.2 Consistency between Interaction Diagrams, Scenarios and Protocols......Page 111
7.6.3 Consistency of Communication Specifications......Page 112
7.6.4 Consistency between Descriptors and the System Overview Diagram......Page 113
8 Detailed Design: Agents, Capabilities and Processes......Page 115
8.1 Capabilities......Page 116
8.2 Agent Overview Diagrams......Page 118
8.3 Process Specifications......Page 119
8.4 Develop Capability and Process Descriptors......Page 122
9 Detailed Design: Capabilities, Plans and Events......Page 125
9.1 Capability Overview Diagrams......Page 126
9.2 Sub-tasks and Alternative Plans......Page 128
9.2.1 Identifying Context Conditions......Page 129
9.2.2 Coverage and Overlap......Page 131
9.3 Events and Messages......Page 132
9.4 Action and Percept Detailed Design......Page 133
9.6 Develop and Refine Descriptors......Page 134
9.7 Checking for Completeness and Consistency......Page 135
9.7.2 Missing or Redundant Items......Page 136
9.7.3 Consistency between Artifacts......Page 137
9.7.4 Important Scenarios......Page 139
10.1 Agent Platforms......Page 141
10.3 Example......Page 143
10.3.1 Agents......Page 144
10.3.2 Capabilities......Page 146
10.3.3 Data......Page 149
10.3.4 Messages/Events......Page 150
10.3.5 Plans......Page 152
10.4 Automatic Generation of Skeleton Code......Page 154
A Electronic Bookstore......Page 155
B Descriptor Forms......Page 215
C The AUML Notation......Page 221
Bibliography......Page 231
Index......Page 237