代理人系統是自動化協商常見的作法,基於代理人本身的特性,在許多生產規劃排程的相關研究中,常會以多代理人之間的協調機制來做為資源衝突的解決方法。代理人之間的協調機制是指多個代理人之間能夠透過合作或協商的方式來協調彼此之間的行動,而代理人之間在互動的過程中,必須使用相同的溝通語言以及互動規則,彼此之間對於溝通的內容還要具有共同的認知,但是在開放式的環境之下,代理人溝通的對象隨時都有可能會改變,而且每個代理人對相同領域的事物都可能會有不同的認知,因此當以往的代理人互動對象愈多的時候,代理人系統就會愈複雜。知識本體可以將代理人在協商過程中所需具備的知識以抽象化的方式來表達,讓代理人之間對溝通的內容能夠有共同的認知,因此本研究將語意網的相關技術應用於代理人的協商,提出一個以知識本體為基礎的代理人協商環境,將生產排程以及協商相關知識以知識本體的方式來表達,做為代理人在協商過程中所需具備的知識,並且利用規則語言來描述互動的規則,希望能夠以知識本體來解決代理人對知識認知的問題,使代理人系統在開放式的環境下能夠和不特定的對象進行協商,以提升代理人系統的彈性。
In this research we present our experience in applying Semantic Web technology to multi-agent negotiation for production scheduling. This result is an ontology-based negotiation environment for multi-agent, which is particularly suitable to open environments such as the Internet. In this environment, agent communication is based on the assumption that two agents, who wish to converse, share a common ontology for the domain of discourse. It ensures that the agents ascribe the same meaning to the symbols used in the message. This research describes a novel approach to negotiation, where the negotiation protocol doesn’t need to be hard-coded in agents, but it is represented by ontology: an explicit and declarative representation of the negotiation protocol. And we use rule language to represent the interaction rules that agent has to follow in order to participate to a negotiation. We use ontology to represent the domain knowledge of production scheduling and negotiation protocol for agents. The ontology of scheduling and negotiation protocol is common understanding of agents. Because the negotiation protocol ontology can be used to represent interaction rules. The advantages of our research are of twofold. The first advantage is flexibility. Agents can change different interaction rule for different negotiation participant. The second advantage is that the ontology provides the terminology to reason in terms of negotiation protocols.
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。