Integrating General Purpose Planners and Specialized Reasoners: Case Study of a Hybrid Planning Architecture

Many real-world planning problems involve sub stantial amounts of domain-specific reasoning that is either awkward or inefficient to encode in a general purpose planner. Previous approaches for planning in such domains have either been largely domain specific or have compromised with shallow models of the domain-specific considerations. In this paper, we propose a hybrid planning architecture for such domains, which utilizes a set of specialists to complement both the overall expressiveness and the efficiency of a traditional hierarchical planner. Such an architecture promises to retain the flexibility and generality of classical planning framework while allowing deeper and more efficient domain-specific reasoning through specialists. The architecture, however, has several ramifications on the internal operations of the planner as well as its interactions with the specialists. First, continual interactions between the planner and the specialists necessitate an incremental, interactive, and least-commitment oriented approach to planning. Second, as the planner and the specialists in such a model may employ heterogeneous reasoning mechanisms and representations, a complete understanding of the operations of one by the other is not possible. This necessitates designing interfaces at the right level of abstraction, to efficiently mediate the interactions between them. In this paper, we investigate these issues with the help of our implementation of a hybrid planning architecture for a manufacturing planning domain.