Publication at Faculty of Mathematics and Physics |
2016
Abstract
Domain-independent planning requires only to specify planning problems in a standard language (e.g. PDDL) in order to utilise planning in some application.
Despite a huge advancement in domain-independent planning, some relatively-easy problems are still challenging for existing planning engines. Such an issue can be mitigated by specifying Domain Control Knowledge (DCK) that can provide better guidance for planning engines.
In this paper, we introduce transition-based DCK, inspired by Finite State Automata, that is efficient as demonstrated empirically, planner-independent (can be encoded within planning problems) and easy to specify.