استفاده از تکنیک های کمی برای حل تعارض در یک سیستم چند عامله

The distributed artificial intelligence (DAI) can create multiple interacting systems and deal with the description. It looks alike that a group of experts’ cooperating together to solve a global problem, which is difficult and decomposed. Each expert can be implemented as an individual agent, which is a self-contained, independent, and active object that can communicate with other agents either indirectly or directly. For achieving a solution to a certain global problem, each agent has its own local goal to be fulfilled. As each agent is given control over its own actions, its own local goals, and its own utility functions, it will naturally pursue its own interests. When working on a shared goal each agent would like its priorities to be given the most consideration. Unfortunately, many agents’ priorities (arising from their own personal goals) will not be coherent with their fellow agents’, thus leading to the conflict phenomenon. Thus, before a satisfactory solution to the whole problem can be presented, this situation of conflict must be resolved. In this paper, we will develop a mathematical model of which tries to resolve the conflict in a multi-agent system by providing a possible solution to the problem. We concentrate on the linear programming model to develop its conflict resolution algorithm and implement it on AGENT-0. Finally, the linear programming algorithm was implemented on the Unix workstations within the AGENT-0 environment that is set under the common Lisp environment. Besides, we evaluated the whole system and proposed some suggestions for future work.

The multi-agent system is the subject of a stream of research within distributed artificial intelligence (DAI). This section will introduce some general ideas about AI and DAI, and then begin our paper field. The AI is a scientific discipline, which resembles physics in the sense that it is interested in behavior and tries to extract models able to reproduce that behavior and explain it. In AI [Gas87], the objects of study are intelligent entities rather than physical objects. The main part of any artificial intelligence application is knowledge, a knowledge which it acquires from human experts and which has many components such as relationships, procedures, facts, theories and concepts. This knowledge is contained within a knowledge base that typically focuses on a specific subject area or domain. The DAI is a sub-area of AI dealing with the description and creation of multiple interacting systems. It is concerned with a set of loosely coupled intelligent agents who will spend more time on computation than on communication between each other, cooperating to solve a global problem. A DAI system is that of a group of experts cooperating together to solve the problems, which are difficult and decomposed [24]. Therefore DAI applies both AI techniques and multiple problem solvers to solve these problems. Each expert is implemented as an agent, which is a self-contained, independent, active object, which can communicate with other agents either directly or indirectly. Thus, DAI is actually interested in the cooperative solution of problems by a decentralized group of agents. This followings presents a brief outline of how this paper has been structured. Section 1 introduces the definitions of both AI and DAI. Section 2 describes the basic ideas of DAI and justify for DAI. Section 3 discusses types of conflict and their associated conflict resolution strategies. It also presents the importance of conflict resolution and previous work carried out in this area. Section 4 describes a certain mathematical formula and techniques to be used. Section 5 gives a brief assessment of this method mentioned in Section 4 and then concentrates on this method to develop its algorithm. Section 6 introduces the idea of ‘AGENT-0’ and describes the design of the programming. Section 7 sees the implementation of the formulae and technique into the algorithm I have developed. Section 8 evaluates the system and describes the future work. Section 9 presents my conclusions of the paper.

The aim of this paper was to develop a generic mathematical model to help resolve conflict in DAI problems. One modest model, linear programming model, was adopted and an algorithm developed and applied in a software language called AGENT-0. This system ought to be viewed as a prototype for other mathematical models in future work. This approach was based on linear programming theory, that is to say, the objective and the constraints of the problem must be expressed by linear equations. The algorithm was implemented within the AGENT-0 environment on the Unix workstations using a simple interface. In AGENT-0 agents are viewed as computational entities possessing formal versions of mental state, and in particular formal versions of beliefs, capabilities, commitments, and possibly other mentalistic qualities. Therefore the views of agents were represented by the beliefs, capabilities and commitments. It makes the problem more realistic. There is a specific case accompanied by the algorithm to let readers see the application of this linear programming model in the real world. To conclude this paper, I suggest that these mathematical models are important in the resolution of conflict in DAI. Although at the moment only a linear programming model has been developed and implemented, but it is designed to contain a decision rule database including some techniques of conflict resolution that the CR agent can use to make decisions. Moreover the CR agent can communicate directly with the individual agents to check whether the result of conflict resolution is satisfied with each agent’s individual constraint or not, in order to increase the accuracy of the result. Thus it can provide the necessary infrastructure into which additional mathematical models could subsequently be accommodated.