Task planning in robotics: an empirical comparison of PDDL- and ASP-based systems
- 73 Downloads
Robots need task planning algorithms to sequence actions toward accomplishing goals that are impossible through individual actions. Off-the-shelf task planners can be used by intelligent robotics practitioners to solve a variety of planning problems. However, many different planners exist, each with different strengths and weaknesses, and there are no general rules for which planner would be best to apply to a given problem. In this study, we empirically compare the performance of state-of-the-art planners that use either the planning domain description language (PDDL) or answer set programming (ASP) as the underlying action language. PDDL is designed for task planning, and PDDL-based planners are widely used for a variety of planning problems. ASP is designed for knowledge-intensive reasoning, but can also be used to solve task planning problems. Given domain encodings that are as similar as possible, we find that PDDL-based planners perform better on problems with longer solutions, and ASP-based planners are better on tasks with a large number of objects or tasks in which complex reasoning is required to reason about action preconditions and effects. The resulting analysis can inform selection among general-purpose planning systems for particular robot task planning domains.
Key wordsTask planning Robotics Planning domain description language (PDDL) Answer set programming (ASP)
Unable to display preview. Download preview PDF.
A portion of this work has taken place in the Learning Agents Research Group (LARG) at UT Austin. LARG research is supported in part by NSF (IIS-1637736, IIS-1651089, IIS-1724157), ONR (N00014-18-2243), FLI (RFP2-000), Intel, Raytheon, and Lockheed Martin. Peter STONE serves on the Board of Directors of Cogitai, Inc. The terms of this arrangement have been reviewed and approved by the University of Texas at Austin in accordance with its policy on objectivity in research.
- Babb J, Lee J, 2015. Action language BC+: preliminary report. Proc 29th AAAI Conf on Artificial Intelligence, p.1424–1430.Google Scholar
- Chen XP, Ji JM, Jiang JQ, et al., 2010. Developing high-level cognitive functions for service robots. Proc 9th Int Conf on Autonomous Agents and Multiagent Systems, p.989–996.Google Scholar
- de la Rosa T, Olaya AG, Borrajo D, 2007. Using cases utility for heuristic planning improvement. Int Conf on Case-Based Reasoning, p.137–148. https://doi.org/10.1007/978-3-540-74141-1_10
- Fawcett C, Vallati M, Hutter F, et al., 2014. Improved features for runtime prediction of domain-independent planners. Proc 24th Int Conf on Automated Planning and Scheduling, p.355–359.Google Scholar
- Gebser M, Grote T, Schaub T, 2010. Coala: a compiler from action languages to ASP. European Workshop on Logics in Artificial Intelligence, p.360–364. https://doi.org/10.1007/978-3-642-15675-5_32
- Gebser M, Kaminski R, Kaufmann B, et al., 2014. Clingo=ASP+control: preliminary report. https://doi.org/arxiv.org/abs/1405.3694
- Helmert M, Röger G, Karpas E, 2011. Fast downward stone soup: a baseline for building planner portfolios. Int Conf on Automated Planning and Scheduling Workshop on Planning and Learning, p.28–35.Google Scholar
- Hoffmann J, 2001. FF: the fast-forward planning system. AI Mag, 22(3):57–62.Google Scholar
- Lee J, Lifschitz V, Yang F, 2013. Action language BC: preliminary report. Proc 23rd Int Joint Conf on Artificial Intelligence, p.983–989.Google Scholar
- Leyton-Brown K, Nudelman E, Shoham Y, 2002. Learning the empirical hardness of optimization problems: the case of combinatorial auctions. Int Conf on Principles and Practice of Constraint Programming, p.556–572. https://doi.org/10.1007/3-540-46135-3_37
- Lifschitz V, 2008. What is answer set programming? Proc 23rd National Conf on Artificial Intelligence, p.1594–1597.Google Scholar
- Lo SY, Zhang S, Stone P, 2018. PETLON: planning efficiently for task-level-optimal navigation. Proc 17th Conf on Autonomous Agents and Multiagent Systems, p.220–228.Google Scholar
- McDermott D, 2003. The formal semantics of processes in PDDL. Proc ICAPS Workshop on PDDL, p.101–155.Google Scholar
- McDermott D, Ghallab M, Howe A, et al., 1998. PDDL—the planning domain definition language. https://doi.org/www.cs.yale.edu/homes/dvm/
- Miura S, Fukunaga A, 2017. Automatic extraction of axioms for planning. Proc 27th Int Conf on Automated Planning and Scheduling, p.218–227.Google Scholar
- Richter S, Westphal M, Helmert M, 2011. Lama 2008 and 2011. Int Planning Competition, p.117–124.Google Scholar
- Yang F, Khandelwal P, Leonetti M, et al., 2014. Planning in answer set programming while learning action costs for mobile robots. AAAI Spring Symp on Knowledge Representation and Reasoning in Robotics, p. 71–78.Google Scholar
- Zhang S, Yang F, Khandelwal P, et al., 2015. Mobile robot planning using action language BC with an abstraction hierarchy. Proc 13th Int Conf on Logic Programming and Nonmonotonic Reasoning, p. 502–516. https://doi.org/10.1007/978-3-319-23264-5_42