Application for An Architecture to Coordinate Multi-Agent Systems
Save to My Collections
Carrillo, A. (2000). Application for An Architecture to Coordinate Multi-Agent Systems. In Proceedings of WebNet World Conference on the WWW and Internet 2000 (pp. 661-662). Chesapeake, VA: AACE.
Retrieved from http://www.editlib.org/p/6451.
Coordination is the process that manages the interdependencies in to the activities. Communications are based in the detection and response to coordination relations between tasks sets that are known by some agents. Coordination discute how agents can generate communication tasks which help the coordination process, how the agents can thrust in the inferences from receivers to carry out their tasks and, how the agents are in favor to plans and communications to carry out concurrent executions using PGP and GPGP. PGPG and GPGP are sets of coordination mechanisms, independent-domain that use constraints to split a local planner in modules. These constraints give importance to some tasks due they help to no-local tasks can be done and their beginning and ending time can be good. In this article, we use Yubarta framework referenced in (Ucros, 1996) to carry out the coordination process for the distribution of the information in an organization that flows between Auditory department and Systems department areas. This process avoid loss, delay or replication of the information. Coordination process was done in two levels : the first level talks about the agents which involve the system and, the second level talks about the task that each agent carries out. For these levels we use Template Mediator, an hierarchical architecture to coordinate agents that is used to distribute task for each agent and to verify the results of themselves.
Comments & Discussion
Comment on the paper above. You must be registered to participate. Registration is free.