| 摘要 |
<p>A Multiple-Agent Hybrid Control Architecture (MAHCA) uses agents to analyze, design, and implement intelligent control of distributed processes. A single agent can be configured to control a local process and a network of agents can be configured to control more complex distributed processes. Each agent includes a knowledge base builder (22) which supports manual input from users, a planner (24) which generates a statement representing the desired behavior of the system as an existentially quantified logic expression, an inferencer (26) which determines whether this statement is a theorem currently active in the knowledge base (28) and if it does the inferencer generates the current control action schedule, an adapter (30) which replaces or modifies statements that do not follow from the current status of the knowledge base, and a knowledge decoder (32) which supports automatic input from other agents. Multiple agents interact through messages and can be either permanent or temporary.
</p> |
