Initial Results on Generating Macro Actions from a Plan Database for Planning on Autonomous Mobile Robots. Hofmann, T., Niemueller, T., & Lakemeyer, G. In Proceedings of the 27th International Conference on Automated Planning and Scheduling (ICAPS), Pittsburgh, PA, USA, 2017. Paper Poster Presentation Project abstract bibtex Planning in an on-line robotics context has the specific requirement of a short planning duration. A property of typical contemporary scenarios is that (mobile) robots perform similar or even repeating tasks during operation. With these robot domains in mind, we propose database-driven macro planning for STRIPS (DBMP/S) that learns macros - action sequences that frequently appear in plans - from experience for PDDL-based planners. Planning duration is improved over time by off-line processing of seed plans using a scalable database. The approach is indifferent about the specific planner by representing the resulting macros again as actions with preconditions and effects determined based on the actions contained in the macro. For some domains we have used separate planners for learning and execution exploiting their respective strengths. Initial results based on some IPC domains and a logistic robot scenario show significantly improved (over non-macro planners) or slightly better and comparable (to existing macro planners) performance.
@InProceedings {DBMP-STRIPS,
author = {Till Hofmann and Tim Niemueller and Gerhard Lakemeyer},
title = {Initial Results on Generating Macro Actions from a Plan Database for
Planning on Autonomous Mobile Robots},
booktitle = {Proceedings of the 27th International Conference on Automated
Planning and Scheduling (ICAPS)},
url = {https://kbsg.rwth-aachen.de/~hofmann/papers/db-macro-planning-short-proceedings.pdf},
url_Poster = {https://kbsg.rwth-aachen.de/~hofmann/papers/db-macro-planning-short-poster.pdf},
url_Presentation = {https://youtu.be/AOQX0vsioqI},
url_Project = {https://www.fawkesrobotics.org/projects/dbmp-strips/},
year = 2017,
address = {Pittsburgh, PA, USA},
abstract = {Planning in an on-line robotics context has the specific
requirement of a short planning duration. A property of typical
contemporary scenarios is that (mobile) robots perform similar or
even repeating tasks during operation. With these robot domains
in mind, we propose database-driven macro planning for STRIPS
(DBMP/S) that learns macros - action sequences that frequently
appear in plans - from experience for PDDL-based planners.
Planning duration is improved over time by off-line processing of
seed plans using a scalable database. The approach is indifferent
about the specific planner by representing the resulting macros
again as actions with preconditions and effects determined based
on the actions contained in the macro. For some domains we have
used separate planners for learning and execution exploiting their
respective strengths. Initial results based on some IPC domains
and a logistic robot scenario show significantly improved (over
non-macro planners) or slightly better and comparable (to existing
macro planners) performance.}
}
Downloads: 0
{"_id":"mzCxyrNySaL8Cb3p7","bibbaseid":"hofmann-niemueller-lakemeyer-initialresultsongeneratingmacroactionsfromaplandatabaseforplanningonautonomousmobilerobots-2017","author_short":["Hofmann, T.","Niemueller, T.","Lakemeyer, G."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["Till"],"propositions":[],"lastnames":["Hofmann"],"suffixes":[]},{"firstnames":["Tim"],"propositions":[],"lastnames":["Niemueller"],"suffixes":[]},{"firstnames":["Gerhard"],"propositions":[],"lastnames":["Lakemeyer"],"suffixes":[]}],"title":"Initial Results on Generating Macro Actions from a Plan Database for Planning on Autonomous Mobile Robots","booktitle":"Proceedings of the 27th International Conference on Automated Planning and Scheduling (ICAPS)","url":"https://kbsg.rwth-aachen.de/~hofmann/papers/db-macro-planning-short-proceedings.pdf","url_poster":"https://kbsg.rwth-aachen.de/~hofmann/papers/db-macro-planning-short-poster.pdf","url_presentation":"https://youtu.be/AOQX0vsioqI","url_project":"https://www.fawkesrobotics.org/projects/dbmp-strips/","year":"2017","address":"Pittsburgh, PA, USA","abstract":"Planning in an on-line robotics context has the specific requirement of a short planning duration. A property of typical contemporary scenarios is that (mobile) robots perform similar or even repeating tasks during operation. With these robot domains in mind, we propose database-driven macro planning for STRIPS (DBMP/S) that learns macros - action sequences that frequently appear in plans - from experience for PDDL-based planners. Planning duration is improved over time by off-line processing of seed plans using a scalable database. The approach is indifferent about the specific planner by representing the resulting macros again as actions with preconditions and effects determined based on the actions contained in the macro. For some domains we have used separate planners for learning and execution exploiting their respective strengths. Initial results based on some IPC domains and a logistic robot scenario show significantly improved (over non-macro planners) or slightly better and comparable (to existing macro planners) performance.","bibtex":"@InProceedings {DBMP-STRIPS,\n author = {Till Hofmann and Tim Niemueller and Gerhard Lakemeyer},\n title = {Initial Results on Generating Macro Actions from a Plan Database for\n Planning on Autonomous Mobile Robots},\n booktitle = {Proceedings of the 27th International Conference on Automated\n Planning and Scheduling (ICAPS)},\n url = {https://kbsg.rwth-aachen.de/~hofmann/papers/db-macro-planning-short-proceedings.pdf},\n url_Poster = {https://kbsg.rwth-aachen.de/~hofmann/papers/db-macro-planning-short-poster.pdf},\n url_Presentation = {https://youtu.be/AOQX0vsioqI},\n url_Project = {https://www.fawkesrobotics.org/projects/dbmp-strips/},\n year = 2017,\n address = {Pittsburgh, PA, USA},\n abstract = {Planning in an on-line robotics context has the specific\n requirement of a short planning duration. A property of typical\n contemporary scenarios is that (mobile) robots perform similar or\n even repeating tasks during operation. With these robot domains\n in mind, we propose database-driven macro planning for STRIPS\n (DBMP/S) that learns macros - action sequences that frequently\n appear in plans - from experience for PDDL-based planners.\n Planning duration is improved over time by off-line processing of\n seed plans using a scalable database. The approach is indifferent\n about the specific planner by representing the resulting macros\n again as actions with preconditions and effects determined based\n on the actions contained in the macro. For some domains we have\n used separate planners for learning and execution exploiting their\n respective strengths. Initial results based on some IPC domains\n and a logistic robot scenario show significantly improved (over\n non-macro planners) or slightly better and comparable (to existing\n macro planners) performance.}\n}\n\n","author_short":["Hofmann, T.","Niemueller, T.","Lakemeyer, G."],"key":"DBMP-STRIPS","id":"DBMP-STRIPS","bibbaseid":"hofmann-niemueller-lakemeyer-initialresultsongeneratingmacroactionsfromaplandatabaseforplanningonautonomousmobilerobots-2017","role":"author","urls":{"Paper":"https://kbsg.rwth-aachen.de/~hofmann/papers/db-macro-planning-short-proceedings.pdf"," poster":"https://kbsg.rwth-aachen.de/~hofmann/papers/db-macro-planning-short-poster.pdf"," presentation":"https://youtu.be/AOQX0vsioqI"," project":"https://www.fawkesrobotics.org/projects/dbmp-strips/"},"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://kbsg.rwth-aachen.de/files/kbsgweb.bib","dataSources":["dqRQPSg6Hy3ZXQg7z"],"keywords":[],"search_terms":["initial","results","generating","macro","actions","plan","database","planning","autonomous","mobile","robots","hofmann","niemueller","lakemeyer"],"title":"Initial Results on Generating Macro Actions from a Plan Database for Planning on Autonomous Mobile Robots","year":2017}