This paper reports the outcome of the third in the series of biennialinternational planning competitions, held in association with the InternationalConference on AI Planning and Scheduling (AIPS) in 2002. In addition todescribing the domains, the planners and the objectives of the competition, thepaper includes analysis of the results. The results are analysed from severalperspectives, in order to address the questions of comparative performancebetween planners, comparative difficulty of domains, the degree of agreementbetween planners about the relative difficulty of individual problem instancesand the question of how well planners scale relative to one another overincreasingly difficult problems. The paper addresses these questions throughstatistical analysis of the raw results of the competition, in order todetermine which results can be considered to be adequately supported by thedata. The paper concludes with a discussion of some challenges for the futureof the competition series.
展开▼
