Automatic acquisition of task theories for robotic manipulation:
Abstract: "A task theory is a collection of models that a robot uses to predict the effects of its actions. This thesis addresses how a robot can build a task theory from observed action effects, thus bypassing the usual human-programming method of defining 'what actions do', and maki...
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Buch |
| Sprache: | English |
| Veröffentlicht: |
Pittsburgh, Pa.
1992
|
| Schlagworte: | |
| Zusammenfassung: | Abstract: "A task theory is a collection of models that a robot uses to predict the effects of its actions. This thesis addresses how a robot can build a task theory from observed action effects, thus bypassing the usual human-programming method of defining 'what actions do', and making robots considerably easier to train. The question that this thesis addresses is: 'What are sufficient conditions whereby a robotic manipulation system can learn action models that allow automated planning and successful execution of manipulation strategies?' This thesis, by presenting an implemented learning robot system, shows that acquiring action models can be automated, and further shows that the required software mechanisms are simple and few The important characteristics of the learning agent and its environment are: (1) the complexity of the task, (2) the sensory and effectory abilities of the agent, (3) the learning mechanism, (4) the planning mechanism, (5) the experimentation method, and (6) the prior knowledge of the task possessed by the agent. Sufficient conditions for success are demonstrated with empirical results in three manipulation tasks: parallel-jaw grasping, the peg-in-hole problem, and the tray- tilting problem. The successful learning robots for these tasks generate task theories consisting of a set of funnels. Each funnel is a kind of operator that maps a region of the task state-action space to a region of the state space Funnels can be acquired for continuous tasks using very little prior knowledge of the task, but the learning mechanism must be robust in the face of noise and non-determinism. A simple search-based planner suffices for generating manipulation plans, although this planner must reason about the reliabilities associated with each funnel. After demonstrating sufficient conditions for successful acquisition of task theories, the thesis considers the necessity of the characteristics of the implemented learning robots. In addition to sensing, effecting, planning, and learning, the thesis develops requirements on the prior knowledge that must be possessed by the learning robot |
| Beschreibung: | Zugl.: Pittsburgh, Pa., Carnegie Mellon Univ., Diss., 1992 |
| Beschreibung: | VI, 173 S. Ill. u. graph. Darst. |
Internformat
MARC
| LEADER | 00000nam a2200000 c 4500 | ||
|---|---|---|---|
| 001 | BV008028566 | ||
| 003 | DE-604 | ||
| 005 | 00000000000000.0 | ||
| 007 | t | ||
| 008 | 930713s1992 ad|| m||| 00||| eng d | ||
| 035 | |a (OCoLC)26142879 | ||
| 035 | |a (DE-599)BVBBV008028566 | ||
| 040 | |a DE-604 |b ger |e rakddb | ||
| 041 | 0 | |a eng | |
| 082 | 0 | |a 510.7808 |b C28r 92-111 | |
| 084 | |a FER 980d |2 stub | ||
| 084 | |a DAT 815d |2 stub | ||
| 088 | |a CMU CS 92 111 | ||
| 100 | 1 | |a Christiansen, Alan D. |e Verfasser |4 aut | |
| 245 | 1 | 0 | |a Automatic acquisition of task theories for robotic manipulation |
| 246 | 1 | 3 | |a CMU CS 92 111 |
| 264 | 1 | |a Pittsburgh, Pa. |c 1992 | |
| 300 | |a VI, 173 S. |b Ill. u. graph. Darst. | ||
| 336 | |b txt |2 rdacontent | ||
| 337 | |b n |2 rdamedia | ||
| 338 | |b nc |2 rdacarrier | ||
| 500 | |a Zugl.: Pittsburgh, Pa., Carnegie Mellon Univ., Diss., 1992 | ||
| 520 | 3 | |a Abstract: "A task theory is a collection of models that a robot uses to predict the effects of its actions. This thesis addresses how a robot can build a task theory from observed action effects, thus bypassing the usual human-programming method of defining 'what actions do', and making robots considerably easier to train. The question that this thesis addresses is: 'What are sufficient conditions whereby a robotic manipulation system can learn action models that allow automated planning and successful execution of manipulation strategies?' This thesis, by presenting an implemented learning robot system, shows that acquiring action models can be automated, and further shows that the required software mechanisms are simple and few | |
| 520 | 3 | |a The important characteristics of the learning agent and its environment are: (1) the complexity of the task, (2) the sensory and effectory abilities of the agent, (3) the learning mechanism, (4) the planning mechanism, (5) the experimentation method, and (6) the prior knowledge of the task possessed by the agent. Sufficient conditions for success are demonstrated with empirical results in three manipulation tasks: parallel-jaw grasping, the peg-in-hole problem, and the tray- tilting problem. The successful learning robots for these tasks generate task theories consisting of a set of funnels. Each funnel is a kind of operator that maps a region of the task state-action space to a region of the state space | |
| 520 | 3 | |a Funnels can be acquired for continuous tasks using very little prior knowledge of the task, but the learning mechanism must be robust in the face of noise and non-determinism. A simple search-based planner suffices for generating manipulation plans, although this planner must reason about the reliabilities associated with each funnel. After demonstrating sufficient conditions for successful acquisition of task theories, the thesis considers the necessity of the characteristics of the implemented learning robots. In addition to sensing, effecting, planning, and learning, the thesis develops requirements on the prior knowledge that must be possessed by the learning robot | |
| 650 | 4 | |a Machine learning | |
| 650 | 4 | |a Robotics | |
| 650 | 0 | 7 | |a Maschinelles Lernen |0 (DE-588)4193754-5 |2 gnd |9 rswk-swf |
| 650 | 0 | 7 | |a Roboter |0 (DE-588)4050208-9 |2 gnd |9 rswk-swf |
| 655 | 7 | |0 (DE-588)4113937-9 |a Hochschulschrift |2 gnd-content | |
| 689 | 0 | 0 | |a Roboter |0 (DE-588)4050208-9 |D s |
| 689 | 0 | 1 | |a Maschinelles Lernen |0 (DE-588)4193754-5 |D s |
| 689 | 0 | |5 DE-604 | |
| 999 | |a oai:aleph.bib-bvb.de:BVB01-005282845 | ||
Datensatz im Suchindex
| _version_ | 1804122392906170368 |
|---|---|
| any_adam_object | |
| author | Christiansen, Alan D. |
| author_facet | Christiansen, Alan D. |
| author_role | aut |
| author_sort | Christiansen, Alan D. |
| author_variant | a d c ad adc |
| building | Verbundindex |
| bvnumber | BV008028566 |
| classification_tum | FER 980d DAT 815d |
| ctrlnum | (OCoLC)26142879 (DE-599)BVBBV008028566 |
| dewey-full | 510.7808 |
| dewey-hundreds | 500 - Natural sciences and mathematics |
| dewey-ones | 510 - Mathematics |
| dewey-raw | 510.7808 |
| dewey-search | 510.7808 |
| dewey-sort | 3510.7808 |
| dewey-tens | 510 - Mathematics |
| discipline | Informatik Mathematik Fertigungstechnik |
| format | Book |
| fullrecord | <?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>03451nam a2200433 c 4500</leader><controlfield tag="001">BV008028566</controlfield><controlfield tag="003">DE-604</controlfield><controlfield tag="005">00000000000000.0</controlfield><controlfield tag="007">t</controlfield><controlfield tag="008">930713s1992 ad|| m||| 00||| eng d</controlfield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(OCoLC)26142879</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)BVBBV008028566</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-604</subfield><subfield code="b">ger</subfield><subfield code="e">rakddb</subfield></datafield><datafield tag="041" ind1="0" ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2=" "><subfield code="a">510.7808</subfield><subfield code="b">C28r 92-111</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">FER 980d</subfield><subfield code="2">stub</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">DAT 815d</subfield><subfield code="2">stub</subfield></datafield><datafield tag="088" ind1=" " ind2=" "><subfield code="a">CMU CS 92 111</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Christiansen, Alan D.</subfield><subfield code="e">Verfasser</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Automatic acquisition of task theories for robotic manipulation</subfield></datafield><datafield tag="246" ind1="1" ind2="3"><subfield code="a">CMU CS 92 111</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="a">Pittsburgh, Pa.</subfield><subfield code="c">1992</subfield></datafield><datafield tag="300" ind1=" " ind2=" "><subfield code="a">VI, 173 S.</subfield><subfield code="b">Ill. u. graph. Darst.</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="b">n</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="b">nc</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="500" ind1=" " ind2=" "><subfield code="a">Zugl.: Pittsburgh, Pa., Carnegie Mellon Univ., Diss., 1992</subfield></datafield><datafield tag="520" ind1="3" ind2=" "><subfield code="a">Abstract: "A task theory is a collection of models that a robot uses to predict the effects of its actions. This thesis addresses how a robot can build a task theory from observed action effects, thus bypassing the usual human-programming method of defining 'what actions do', and making robots considerably easier to train. The question that this thesis addresses is: 'What are sufficient conditions whereby a robotic manipulation system can learn action models that allow automated planning and successful execution of manipulation strategies?' This thesis, by presenting an implemented learning robot system, shows that acquiring action models can be automated, and further shows that the required software mechanisms are simple and few</subfield></datafield><datafield tag="520" ind1="3" ind2=" "><subfield code="a">The important characteristics of the learning agent and its environment are: (1) the complexity of the task, (2) the sensory and effectory abilities of the agent, (3) the learning mechanism, (4) the planning mechanism, (5) the experimentation method, and (6) the prior knowledge of the task possessed by the agent. Sufficient conditions for success are demonstrated with empirical results in three manipulation tasks: parallel-jaw grasping, the peg-in-hole problem, and the tray- tilting problem. The successful learning robots for these tasks generate task theories consisting of a set of funnels. Each funnel is a kind of operator that maps a region of the task state-action space to a region of the state space</subfield></datafield><datafield tag="520" ind1="3" ind2=" "><subfield code="a">Funnels can be acquired for continuous tasks using very little prior knowledge of the task, but the learning mechanism must be robust in the face of noise and non-determinism. A simple search-based planner suffices for generating manipulation plans, although this planner must reason about the reliabilities associated with each funnel. After demonstrating sufficient conditions for successful acquisition of task theories, the thesis considers the necessity of the characteristics of the implemented learning robots. In addition to sensing, effecting, planning, and learning, the thesis develops requirements on the prior knowledge that must be possessed by the learning robot</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Machine learning</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Robotics</subfield></datafield><datafield tag="650" ind1="0" ind2="7"><subfield code="a">Maschinelles Lernen</subfield><subfield code="0">(DE-588)4193754-5</subfield><subfield code="2">gnd</subfield><subfield code="9">rswk-swf</subfield></datafield><datafield tag="650" ind1="0" ind2="7"><subfield code="a">Roboter</subfield><subfield code="0">(DE-588)4050208-9</subfield><subfield code="2">gnd</subfield><subfield code="9">rswk-swf</subfield></datafield><datafield tag="655" ind1=" " ind2="7"><subfield code="0">(DE-588)4113937-9</subfield><subfield code="a">Hochschulschrift</subfield><subfield code="2">gnd-content</subfield></datafield><datafield tag="689" ind1="0" ind2="0"><subfield code="a">Roboter</subfield><subfield code="0">(DE-588)4050208-9</subfield><subfield code="D">s</subfield></datafield><datafield tag="689" ind1="0" ind2="1"><subfield code="a">Maschinelles Lernen</subfield><subfield code="0">(DE-588)4193754-5</subfield><subfield code="D">s</subfield></datafield><datafield tag="689" ind1="0" ind2=" "><subfield code="5">DE-604</subfield></datafield><datafield tag="999" ind1=" " ind2=" "><subfield code="a">oai:aleph.bib-bvb.de:BVB01-005282845</subfield></datafield></record></collection> |
| genre | (DE-588)4113937-9 Hochschulschrift gnd-content |
| genre_facet | Hochschulschrift |
| id | DE-604.BV008028566 |
| illustrated | Illustrated |
| indexdate | 2024-07-09T17:13:09Z |
| institution | BVB |
| language | English |
| oai_aleph_id | oai:aleph.bib-bvb.de:BVB01-005282845 |
| oclc_num | 26142879 |
| open_access_boolean | |
| physical | VI, 173 S. Ill. u. graph. Darst. |
| publishDate | 1992 |
| publishDateSearch | 1992 |
| publishDateSort | 1992 |
| record_format | marc |
| spelling | Christiansen, Alan D. Verfasser aut Automatic acquisition of task theories for robotic manipulation CMU CS 92 111 Pittsburgh, Pa. 1992 VI, 173 S. Ill. u. graph. Darst. txt rdacontent n rdamedia nc rdacarrier Zugl.: Pittsburgh, Pa., Carnegie Mellon Univ., Diss., 1992 Abstract: "A task theory is a collection of models that a robot uses to predict the effects of its actions. This thesis addresses how a robot can build a task theory from observed action effects, thus bypassing the usual human-programming method of defining 'what actions do', and making robots considerably easier to train. The question that this thesis addresses is: 'What are sufficient conditions whereby a robotic manipulation system can learn action models that allow automated planning and successful execution of manipulation strategies?' This thesis, by presenting an implemented learning robot system, shows that acquiring action models can be automated, and further shows that the required software mechanisms are simple and few The important characteristics of the learning agent and its environment are: (1) the complexity of the task, (2) the sensory and effectory abilities of the agent, (3) the learning mechanism, (4) the planning mechanism, (5) the experimentation method, and (6) the prior knowledge of the task possessed by the agent. Sufficient conditions for success are demonstrated with empirical results in three manipulation tasks: parallel-jaw grasping, the peg-in-hole problem, and the tray- tilting problem. The successful learning robots for these tasks generate task theories consisting of a set of funnels. Each funnel is a kind of operator that maps a region of the task state-action space to a region of the state space Funnels can be acquired for continuous tasks using very little prior knowledge of the task, but the learning mechanism must be robust in the face of noise and non-determinism. A simple search-based planner suffices for generating manipulation plans, although this planner must reason about the reliabilities associated with each funnel. After demonstrating sufficient conditions for successful acquisition of task theories, the thesis considers the necessity of the characteristics of the implemented learning robots. In addition to sensing, effecting, planning, and learning, the thesis develops requirements on the prior knowledge that must be possessed by the learning robot Machine learning Robotics Maschinelles Lernen (DE-588)4193754-5 gnd rswk-swf Roboter (DE-588)4050208-9 gnd rswk-swf (DE-588)4113937-9 Hochschulschrift gnd-content Roboter (DE-588)4050208-9 s Maschinelles Lernen (DE-588)4193754-5 s DE-604 |
| spellingShingle | Christiansen, Alan D. Automatic acquisition of task theories for robotic manipulation Machine learning Robotics Maschinelles Lernen (DE-588)4193754-5 gnd Roboter (DE-588)4050208-9 gnd |
| subject_GND | (DE-588)4193754-5 (DE-588)4050208-9 (DE-588)4113937-9 |
| title | Automatic acquisition of task theories for robotic manipulation |
| title_alt | CMU CS 92 111 |
| title_auth | Automatic acquisition of task theories for robotic manipulation |
| title_exact_search | Automatic acquisition of task theories for robotic manipulation |
| title_full | Automatic acquisition of task theories for robotic manipulation |
| title_fullStr | Automatic acquisition of task theories for robotic manipulation |
| title_full_unstemmed | Automatic acquisition of task theories for robotic manipulation |
| title_short | Automatic acquisition of task theories for robotic manipulation |
| title_sort | automatic acquisition of task theories for robotic manipulation |
| topic | Machine learning Robotics Maschinelles Lernen (DE-588)4193754-5 gnd Roboter (DE-588)4050208-9 gnd |
| topic_facet | Machine learning Robotics Maschinelles Lernen Roboter Hochschulschrift |
| work_keys_str_mv | AT christiansenaland automaticacquisitionoftasktheoriesforroboticmanipulation AT christiansenaland cmucs92111 |