Proceedings of the Sixth International Conference on Artificial Intelligence Planning Systems
AI planning and scheduling has made considerable progress in terms of the theoretical understanding of the fundamental problems, of different planning methods, of algorithms and systems, and of applications of all these. Diversity is still characteristic for the field, and it shows in this proceedings volume. Clustering the state-of-the-art technical papers for AIPS 2002 was, as would be expected, a hairy and somewhat subjective matter as scientific progress seldom respects neat a-priori taxonomies. The clusters that we have formed from the 32 accepted out of 92 submitted papers reflect nicely this diversity, the wide spectrum as well as the deepening of the field when compared to 1992:
- Methods, representations and algorithms for planning are examined in different varieties.
- Analytical, theoretical, and empirical studies have been developed for many of these varieties.
- Quite a number of approaches are motivated by applications of very diverse types.
Many of the papers are focused on or include in some form temporal planning or scheduling. The latter point is particularly interesting. The bonds that should ideally exist between AI-based plan synthesis and OR-based task scheduling have been pointed out since long, but not much technical work has been done in the past to make them real, although it is obvious that their combination leads to highly relevant approaches and systems. Now, temporal planning comes into focus in various forms. The development of a temporal extension of the PDDL planning domain description language, used for the AIPS 2002 planning system contest, is an illustration of these concerns. So, after all these years, AI planning and scheduling is a lively research area. This impression was underlined at the conference by the presence of a large percentage of graduate and PhD students among the participants, including authors or co-authors of accepted papers. Generous grants for scholarships from PLANET, the European network of excellence for planning, from AAAI, NASA, and RIACS have helped make possible these participations.