BitBlast

Benchmarks

We included all benchmarks we used in our experiments in the folder benchmark/.

Note that the domains MPrime, Sugar and Pathways have been grounded beforehand to deal with undefined numeric variables in the initial state, which are not supported by our compilations.

Also, note that some domains, such as HPower, originally contain rational numbers. Variables and constants of these domains have been rescaled to integers when necessary.

Installation

We suggest to start with a new Python environment. We tested the compiler with Python 3.10.

Install the necessary dependencies

pip install unified-planning click sympy --no-cache-dir

Switch to the "BitBlast" directory and install the library.

cd BitBlast
pip install .

How to use Blast, Blast_x and OneHot

The main entrypoint of the compiler is bin/compile.py. To run the compilations use:

python ./bin/compile.py DOMAIN PROBLEM OUTPUT --bits k --COMPILATION

Where:

• DOMAIN is the path to the pddl domain file.
• PROBLEM is the path to the pddl problem file.
• OUTPUT is the path to the output folder (that must exist).
• k is the number of bits to be used by the compilation.
• --COMPILATION specifies the compilation scheme among --base for Blast, --axioms for Blast_x, and --unitary for OneHot.

Examples

To compile the first instance of Counters using Blast with 8 bits run

python ./bin/compile.py benchmark/Counters/instances/domain.pddl benchmark/Counters/instances/pfile1.pddl . --bits 8 --base

To compile the same instance with Blast_x run

python ./bin/compile.py benchmark/Counters/instances/domain.pddl benchmark/Counters/instances/pfile1.pddl . --bits 8 --axioms

To use OneHot with the numeric domain D = [-128, 127] run

python ./bin/compile.py benchmark/Counters/instances/domain.pddl benchmark/Counters/instances/pfile1.pddl . --bits 8 --unitary

How to cite our work

@inproceedings{DBLP:conf/aaai/BonassiPS25,
  author       = {Luigi Bonassi and
                  Francesco Percassi and
                  Enrico Scala},
  title        = {Towards Practical Classical Planning Compilations of Numeric Planning},
  booktitle    = {{AAAI}},
  pages        = {26472--26480},
  publisher    = {{AAAI} Press},
  year         = {2025}
}