Using Joint Intentions in Electricity Transport Management
Nicholas R. Jennings · ePrints Soton (University of Southampton) · 1992
There have been few serious attempts at applying Distributed AI (DAI) techniques to real size, industrial problems. One of the major stumbling blocks to this advancement has been the lack of a clear, implementable theory describing how groups of agents should interact during collaborative problem solving. Such a theory becomes especially important in complex domains in which events occur at unpredictable times, in which decisions are based on incomplete and imprecise information, in which agents possess multiple areas of problem solving competence and when social interactions are complex (i.e. involve several iterations over a prolonged period of time). In such environments it is difficult to ensure that a group’s behaviour remains coordinated, because initial assumptions and deductions may be incorrect or inappropriate; therefore a comprehensive theory must provide a grounded basis from which robust problem solving communities can be constructed.