Adapting Everyday Manipulation Skills to Varied Scenarios

Pawel Gajewski; Paulo Ferreira; Georg Bartels; Chaozheng Wang; Frank Guerin; Bipin Indurkhya; Michael Beetz; Bartlomiej Sniezynski

Adapting Everyday Manipulation Skills to Varied Scenarios

Pawel Gajewski, Paulo Ferreira, Georg Bartels, Chaozheng Wang, Frank Guerin, Bipin Indurkhya, Michael Beetz, Bartlomiej Sniezynski

Research output: Chapter in Book/Report/Conference proceeding › Published conference contribution

15 Citations (Scopus)

5 Downloads (Pure)

Abstract

We address the problem of executing tool-using manipulation skills in scenarios where the objects to be used may vary. We assume that point clouds of the tool and target object can be obtained, but no interpretation or further knowledge about these objects is provided. The system must interpret the point clouds and decide how to use the tool to complete a manipulation task with a target object; this means it must adjust motion trajectories appropriately to complete the task. We tackle three everyday manipulations: scraping material from a tool into a container, cutting, and scooping from a container. Our solution encodes these manipulation skills in a generic way, with parameters that can be filled in at run-time via queries to a robot perception module; the perception module abstracts the functional parts for the tool and extracts key parameters that are needed for the task. The approach is evaluated in simulation and with selected examples on a PR2 robot.

Original language	English
Title of host publication	Proceedings 2019 International Conference on Robotics and Automation (ICRA)
Publisher	IEEE Explore
Publication status	Accepted/In press - 26 Jan 2019
Event	2019 International Conference on Robotics and Automation (ICRA) - Palais des congres de Montreal, Montreal, Canada Duration: 20 May 2019 → 24 May 2019

Conference

Conference	2019 International Conference on Robotics and Automation (ICRA)
Country/Territory	Canada
City	Montreal
Period	20/05/19 → 24/05/19

Bibliographical note

This work is partially funded by: (1) AGH University of Science and Technology, grant No 15.11.230.318. (2) Deutsche Forschungsgemeinschaft (DFG) through the Collaborative Research Center 1320, EASE. (3) Elphinstone Scholarship from University of Aberdeen.

Keywords

cs.RO

Access to Document

Accepted ManuscriptAccepted author manuscript, 2.44 MBLicence: Unspecified

https://arxiv.org/abs/1803.02743Licence: Other

Cite this

Gajewski, P, Ferreira, P, Bartels, G, Wang, C, Guerin, F, Indurkhya, B, Beetz, M & Sniezynski, B 2019, Adapting Everyday Manipulation Skills to Varied Scenarios. in Proceedings 2019 International Conference on Robotics and Automation (ICRA)., 1245, IEEE Explore, 2019 International Conference on Robotics and Automation (ICRA), Montreal, Canada, 20/05/19. <https://arxiv.org/abs/1803.02743>

@inproceedings{16cd11878ea64ec6af9cb2345a7d94d6,

title = "Adapting Everyday Manipulation Skills to Varied Scenarios",

abstract = "We address the problem of executing tool-using manipulation skills in scenarios where the objects to be used may vary. We assume that point clouds of the tool and target object can be obtained, but no interpretation or further knowledge about these objects is provided. The system must interpret the point clouds and decide how to use the tool to complete a manipulation task with a target object; this means it must adjust motion trajectories appropriately to complete the task. We tackle three everyday manipulations: scraping material from a tool into a container, cutting, and scooping from a container. Our solution encodes these manipulation skills in a generic way, with parameters that can be filled in at run-time via queries to a robot perception module; the perception module abstracts the functional parts for the tool and extracts key parameters that are needed for the task. The approach is evaluated in simulation and with selected examples on a PR2 robot.",

keywords = "cs.RO",

author = "Pawel Gajewski and Paulo Ferreira and Georg Bartels and Chaozheng Wang and Frank Guerin and Bipin Indurkhya and Michael Beetz and Bartlomiej Sniezynski",

note = "This work is partially funded by: (1) AGH University of Science and Technology, grant No 15.11.230.318. (2) Deutsche Forschungsgemeinschaft (DFG) through the Collaborative Research Center 1320, EASE. (3) Elphinstone Scholarship from University of Aberdeen.; 2019 International Conference on Robotics and Automation (ICRA) ; Conference date: 20-05-2019 Through 24-05-2019",

year = "2019",

month = jan,

day = "26",

language = "English",

booktitle = "Proceedings 2019 International Conference on Robotics and Automation (ICRA)",

publisher = "IEEE Explore",

}

TY - GEN

T1 - Adapting Everyday Manipulation Skills to Varied Scenarios

AU - Gajewski, Pawel

AU - Ferreira, Paulo

AU - Bartels, Georg

AU - Wang, Chaozheng

AU - Guerin, Frank

AU - Indurkhya, Bipin

AU - Beetz, Michael

AU - Sniezynski, Bartlomiej

N1 - This work is partially funded by: (1) AGH University of Science and Technology, grant No 15.11.230.318. (2) Deutsche Forschungsgemeinschaft (DFG) through the Collaborative Research Center 1320, EASE. (3) Elphinstone Scholarship from University of Aberdeen.

PY - 2019/1/26

Y1 - 2019/1/26

N2 - We address the problem of executing tool-using manipulation skills in scenarios where the objects to be used may vary. We assume that point clouds of the tool and target object can be obtained, but no interpretation or further knowledge about these objects is provided. The system must interpret the point clouds and decide how to use the tool to complete a manipulation task with a target object; this means it must adjust motion trajectories appropriately to complete the task. We tackle three everyday manipulations: scraping material from a tool into a container, cutting, and scooping from a container. Our solution encodes these manipulation skills in a generic way, with parameters that can be filled in at run-time via queries to a robot perception module; the perception module abstracts the functional parts for the tool and extracts key parameters that are needed for the task. The approach is evaluated in simulation and with selected examples on a PR2 robot.

AB - We address the problem of executing tool-using manipulation skills in scenarios where the objects to be used may vary. We assume that point clouds of the tool and target object can be obtained, but no interpretation or further knowledge about these objects is provided. The system must interpret the point clouds and decide how to use the tool to complete a manipulation task with a target object; this means it must adjust motion trajectories appropriately to complete the task. We tackle three everyday manipulations: scraping material from a tool into a container, cutting, and scooping from a container. Our solution encodes these manipulation skills in a generic way, with parameters that can be filled in at run-time via queries to a robot perception module; the perception module abstracts the functional parts for the tool and extracts key parameters that are needed for the task. The approach is evaluated in simulation and with selected examples on a PR2 robot.

KW - cs.RO

UR - http://www.scopus.com/inward/record.url?scp=85071456745&partnerID=8YFLogxK

M3 - Published conference contribution

BT - Proceedings 2019 International Conference on Robotics and Automation (ICRA)

PB - IEEE Explore

T2 - 2019 International Conference on Robotics and Automation (ICRA)

Y2 - 20 May 2019 through 24 May 2019

ER -

Adapting Everyday Manipulation Skills to Varied Scenarios

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this