Housekeeping with Multiple Autonomous Robots: Knowledge Representation and Automated Reasoning for a Tightly Integrated Robot Control Architecture. Aker, E., Erdogan, A., Erdem, E., & Patoglu, V. In Workshop on Knowledge Representation for Autonomous Robots, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2011), 2011. abstract bibtex 2 downloads We embed knowledge representation and automated reasoning in each level of the classical 3-layer robot control architecture, in such a way as to tightly integrate these layers. At the high-level, we represent not only actions and change but also commonsense knowledge in the action description language~p̧lus. Geometric reasoning is lifted to the high-level by embedding motion planning in the domain description, using external predicates. Then a discrete plan is computed for each robot, using the causal reasoner CCALC. At the mid-level, if a continuous trajectory is not computed by a motion planner because the discrete plan is not feasible at the continuous-level, then formal queries are asked to the causal reasoner to find a different plan subject to some (temporal) conditions represented as formulas. At the low-level, if the plan execution fails, then a new continuous trajectory is computed by a motion planner at the mid-level or a new discrete plan is computed using an automated reasoner at the high-level. We apply this tightly integrated robot control architecture in a housekeeping domain with multiple autonomous robots, and illustrate this application with a simulation.
@InProceedings{Aker2011,
booktitle = {Workshop on Knowledge Representation for Autonomous Robots, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2011)},
author = {Erdi Aker and Ahmetcan Erdogan and Esra Erdem and Volkan Patoglu},
title = {Housekeeping with Multiple Autonomous Robots: Knowledge Representation and Automated Reasoning for a Tightly Integrated Robot Control Architecture},
year = {2011},
abstract = {We embed knowledge representation and automated reasoning in each level of the classical 3-layer robot control architecture, in such a
way as to tightly integrate these layers. At the high-level, we represent not only actions and change but also commonsense knowledge
in the action description language~\cplus. Geometric reasoning is lifted to the high-level by embedding motion planning in the domain
description, using external predicates. Then a discrete plan is computed for each robot, using the causal reasoner CCALC. At the
mid-level, if a continuous trajectory is not computed by a motion planner because the discrete plan is not feasible at the
continuous-level, then formal queries are asked to the causal reasoner to find a different plan subject to some (temporal)
conditions represented as formulas. At the low-level, if the plan execution fails, then a new continuous trajectory is computed by a motion planner at the mid-level or a new
discrete plan is computed using an automated reasoner at the high-level. We apply this tightly integrated robot control
architecture in a housekeeping domain with multiple autonomous robots, and illustrate this application with a simulation. }
}
Downloads: 2
{"_id":{"_str":"534213c9ecd21cdc07000070"},"__v":20,"authorIDs":["5456f0758b01c81930000081","5456f7f58b01c819300000d0","545739722abc8e9f370001ce","5477280fcb4bf04e39001677"],"author_short":["Aker, E.","Erdogan, A.","Erdem, E.","Patoglu, V."],"bibbaseid":"aker-erdogan-erdem-patoglu-housekeepingwithmultipleautonomousrobotsknowledgerepresentationandautomatedreasoningforatightlyintegratedrobotcontrolarchitecture-2011","bibdata":{"bibtype":"inproceedings","type":"inproceedings","booktitle":"Workshop on Knowledge Representation for Autonomous Robots, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2011)","author":[{"firstnames":["Erdi"],"propositions":[],"lastnames":["Aker"],"suffixes":[]},{"firstnames":["Ahmetcan"],"propositions":[],"lastnames":["Erdogan"],"suffixes":[]},{"firstnames":["Esra"],"propositions":[],"lastnames":["Erdem"],"suffixes":[]},{"firstnames":["Volkan"],"propositions":[],"lastnames":["Patoglu"],"suffixes":[]}],"title":"Housekeeping with Multiple Autonomous Robots: Knowledge Representation and Automated Reasoning for a Tightly Integrated Robot Control Architecture","year":"2011","abstract":"We embed knowledge representation and automated reasoning in each level of the classical 3-layer robot control architecture, in such a way as to tightly integrate these layers. At the high-level, we represent not only actions and change but also commonsense knowledge in the action description language~p̧lus. Geometric reasoning is lifted to the high-level by embedding motion planning in the domain description, using external predicates. Then a discrete plan is computed for each robot, using the causal reasoner CCALC. At the mid-level, if a continuous trajectory is not computed by a motion planner because the discrete plan is not feasible at the continuous-level, then formal queries are asked to the causal reasoner to find a different plan subject to some (temporal) conditions represented as formulas. At the low-level, if the plan execution fails, then a new continuous trajectory is computed by a motion planner at the mid-level or a new discrete plan is computed using an automated reasoner at the high-level. We apply this tightly integrated robot control architecture in a housekeeping domain with multiple autonomous robots, and illustrate this application with a simulation. ","bibtex":"@InProceedings{Aker2011,\n\tbooktitle = {Workshop on Knowledge Representation for Autonomous Robots, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2011)},\n\tauthor = {Erdi Aker and Ahmetcan Erdogan and Esra Erdem and Volkan Patoglu},\n\ttitle = {Housekeeping with Multiple Autonomous Robots: Knowledge Representation and Automated Reasoning for a Tightly Integrated Robot Control Architecture},\n\tyear = {2011},\n\tabstract = {We embed knowledge representation and automated reasoning in each level of the classical 3-layer robot control architecture, in such a\nway as to tightly integrate these layers. At the high-level, we represent not only actions and change but also commonsense knowledge\nin the action description language~\\cplus. Geometric reasoning is lifted to the high-level by embedding motion planning in the domain\ndescription, using external predicates. Then a discrete plan is computed for each robot, using the causal reasoner CCALC. At the\nmid-level, if a continuous trajectory is not computed by a motion planner because the discrete plan is not feasible at the\ncontinuous-level, then formal queries are asked to the causal reasoner to find a different plan subject to some (temporal)\nconditions represented as formulas. At the low-level, if the plan execution fails, then a new continuous trajectory is computed by a motion planner at the mid-level or a new\ndiscrete plan is computed using an automated reasoner at the high-level. We apply this tightly integrated robot control\narchitecture in a housekeeping domain with multiple autonomous robots, and illustrate this application with a simulation. }\n}\n\n\n","author_short":["Aker, E.","Erdogan, A.","Erdem, E.","Patoglu, V."],"key":"Aker2011","id":"Aker2011","bibbaseid":"aker-erdogan-erdem-patoglu-housekeepingwithmultipleautonomousrobotsknowledgerepresentationandautomatedreasoningforatightlyintegratedrobotcontrolarchitecture-2011","role":"author","urls":{},"downloads":2,"html":"","metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"hmi.sabanciuniv.edu/patoglu_bib.bib","downloads":2,"keywords":[],"search_terms":["housekeeping","multiple","autonomous","robots","knowledge","representation","automated","reasoning","tightly","integrated","robot","control","architecture","aker","erdogan","erdem","patoglu"],"title":"Housekeeping with Multiple Autonomous Robots: Knowledge Representation and Automated Reasoning for a Tightly Integrated Robot Control Architecture","year":2011,"dataSources":["g2mDMKihSyQjpNZsS"]}