Planning under uncertainty for extended reachability goals

Planning under uncertainty has being increasingly demanded for practical applications in several areas that require reliable solutions for complex goals and, therefore, some approaches for automatic plan synthesis based on formal methods have been proposed in AI Planning. Among these approaches, planning based on model checking seems to be a very attractive one, however, it is mainly based on ctl and deals only with simple reachability goals, as those in classical planning. In this paper, we introduce a more expressive class of planning goals (extended reachability goals) and show that, for this class of goals, the ctl’s semantics is not adequate to formalize plan synthesis (or validation) algorithms. Then, as a way to overcome this limitation, we propose a new variant of ctl, based on which we implement plan synthesis (and validation) algorithms capable of solving problems with extended reachability goals, in totally observable nondeterministic environments. Student level: PhD Date of conclusion: November, 2007 International publications from the thesis: [13], [14], [15] and [16]. Link: www.teses.usp.br/teses/disponiveis/45/45134/tde-09042008-105750