Developing Intelligent Agent Systems: A Practical
Guide
Table of Contents
- Contents
- Foreward from the Series Editor (Michael Wooldridge)
- Preface
- Acknowledgements
- Agents and Multi-Agent Systems
- What is an Intelligent Agent?
- Why are Agents Useful?
- Concepts for Building Agents
- Situated Agents: Actions and Percepts
- Proactive and Reactive Agents: Goals and Events
- Challenging Agent Environments: Plans and Beliefs
- Social Agents
- Agent Execution Cycle
- Summary
- Overview of the Prometheus Methodology
- Why a New Methodology?
- Prometheus: A Brief Overview
- Guidelines for Using Prometheus
- Agent-Oriented Methodologies
- System Specification
- Goal Specification
- Functionalities
- Scenario Development
- Interface Description
- Checking for Completeness and Consistency
- Architectural Design: Specifying the Agent Types
- Deciding on the Agent Types
- Grouping functionalities
- Review Agent Coupling - Acquaintance Diagrams
- Develop Agent Descriptors
- Architectural Design: Specifying the Interactions
- Interaction Diagrams from Scenarios
- Interaction Protocols from Interaction Diagrams
- Develop Protocol and Message Descriptors
- Finalizing the Architectural Design
- Overall System Structure
- Identifying Boundaries of the Agent System
- Describing Percepts and Actions
- Defining Shared Data Objects
- System Overview Diagram
- Checking for Completeness and Consistency
- Detailed Design: Agents, Capabilities and Processes
- Capabilities
- Agent Overview Diagrams
- Process Specifications
- Develop Capability and Process Descriptors
- Detailed Design: Capabilities, Plans and Events
- Capability Overview Diagrams
- Sub-tasks and Alternative Plans
- Events and Messages
- Action and Percept Detailed Design
- Data
- Develop and Refine Descriptors
- Checking for Completeness and Consistency
- Implementing Agent Systems
- Agent Platforms
- JACK
- Example
- Automatic Generation of Skeleton Code
- Electronic Bookstore
- Descriptor Forms
- The AUML Notation