Track:
Contents
Downloads:
Abstract:
This work is aboutplan adjustment in the small. A multiagent system is presented that executes plans in an uncertainenvironment. During plan execution environmental changesforce the agents to recognize the deviation, to adapt theiractions to unexpected events and to inform concerned agentsabouthe changes. The agents aim at staying as close as possible to the original plan and to minimize changes. The worksuggests the elements of a plan representation for executingplans in a team in an uncertain environment. It works out towhich deviations a plan can be adjusted without requiringglobal and complete knowledge and without re-planning theactions of several agents. It identifies dependence relationsthat allow to propagate changes to concerned agents.