Code and data for the ICAPS 2022 paper "New Refinement Strategies for Cartesian Abstractions"

Code

The file speck-seipp-icaps2022-code.zip contains an extended version of the Fast Downward planning system (http://fast-downward.org). Please see http://www.fast-downward.org for detailed instructions on how to compile the planner. Here is the short version for building the planner and running the best performing configuration. The calls for the other configurations can be found in the file speck-seipp-icaps2022-reports.zip.

./build.py
./fast-downward.py PDDL_TASK --search "astar(cegar(subtasks=[original()], max_states=infinity, max_transitions=infinity, max_time=900, pick_split=max_cover, tiebreak_split=max_refined, pick_flaw=min_h_batch_multi_split,max_state_expansions=1000000, use_general_costs=true, debug=false, transform=no_transform(), cache_estimates=true, random_seed=-1))"

The latest version of the code is maintained at  https://github.com/jendrikseipp/scorpion.

Benchmarks

The file speck-seipp-icaps2022-benchmarks.zip contains the STRIPS PDDL benchmarks from sequential optimization tracks of IPC 1998-2018.

Experiment data

The remaining zipfiles contain the raw experiment data, parsed values, scripts and basic reports for the experiments in the paper.