SubtaskGenerator

This page describes different subtask (abstraction) generators. They are used by the additive cartesian cegar heuristic .

Abstraction by goals

For each goal atom of the original task one subproblem is generated having only the atom as its goal.

goals(order=hadd_down, random_seed=-1)

• order ({original, random, hadd_up, hadd_down}): ordering of goal or landmark facts
  • original: according to their (internal) variable index
  • random: according to a random permutation
  • hadd_up: according to their h^add value, lowest first
  • hadd_down: according to their h^add value, highest first
• random_seed (int [-1, infinity]): Set to -1 (default) to use the global random number generator. Set to any other value to use a local random number generator with the given seed.

Abstraction by landmarks

For each fact landmark of the delete relaxation of the original task one subproblem is generated having only the landmark as goal. This is a generalization of abstractions by goals.

landmarks(order=hadd_down, random_seed=-1, combine_facts=true)

• order ({original, random, hadd_up, hadd_down}): ordering of goal or landmark facts
  • original: according to their (internal) variable index
  • random: according to a random permutation
  • hadd_up: according to their h^add value, lowest first
  • hadd_down: according to their h^add value, highest first
• random_seed (int [-1, infinity]): Set to -1 (default) to use the global random number generator. Set to any other value to use a local random number generator with the given seed.
• combine_facts (bool): combine landmark facts with domain abstraction

No abstraction

Copies of the original task are used as subproblems.

original(copies=1)

• copies (int [1, infinity]): number of task copies