Several real world applications require planners that deal with  non-deterministic domains and with temporally extended goals. Recent  research is addressing this planning problem. However, the ability  of dealing in practice with large state spaces is still an open  problem. In this paper we describe a planning algorithm for extended goals  that makes use of BDD-based symbolic model checking  techniques. We implement the algorithm in the MBP planner,  evaluate its applicability experimentally, and compare it with  existing tools and algorithms. The results show that, in spite of the difficulty of the problem, MBP deals in practice with domains of large size and with  goals of a certain complexity.

The ability to automatically compose Web Services is critical for realising  more complex functionalities.  Several proposals to use automated planning to deal with the problem of service composition have been recently made. We present an approach, based on modelling the problem as a CSP (Constraint  Satisfaction Problem), that accommodates for the use of numeric variables, sensing and incomplete knowledge. We introduce a language for expressing extended goals, equipped with temporal constructs, maintainability properties, and an explicit distinction between sensing and achievement goals, in order to avoid undesirable situations.