Modeling and learning robot manipulation strategies

Jiming Liu; Y. Y. Tang; Oussama Khatib

doi:10.1007/BFb0113002

Modeling and learning robot manipulation strategies

Jiming Liu
, Y. Y. Tang
, Oussama Khatib

Department of Computer Science

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

Abstract

This paper describes a general approach to learning and planning robot manipulation strategies. Here, the strategies are represented using a discrete-event dynamical systems model where each node corresponds to a state in the robot task environment that triggers certain action schemata and each arc corresponds to a plausible action that brings the task environment into a new state. With such a representation, a manipulation strategy plan can be derived by searching a connected state transition path that is the most reliable. Here, we define the notion of reliability in terms of the estimated chance of success in reaching a desirable state. In the paper, we first present the formalism of discrete-event dynamical system in the context of robot manipulation tasks. Throughout the paper, we provide both illustrative and experimental examples to demonstrate the proposed approach.

Original language	English
Title of host publication	Experimental Robotics V
Subtitle of host publication	The Fifth International Symposium Barcelona, Catalonia, June 15-18, 1997
Editors	Alicia Casals, Anibal T. Almeida
Publisher	Springer Berlin Heidelberg
Chapter	14
Pages	687-700
Number of pages	14
Edition	1
ISBN (Electronic)	9783540409205
DOIs	https://doi.org/10.1007/BFb0113002
Publication status	Published - 15 Jun 1997
Event	5th International Symposium on Experimental Robotics, ISER 1997 - Barcelona, Catalonia, Spain Duration: 15 Jun 1997 → 18 Jun 1997 https://link.springer.com/book/10.1007/BFb0112945 (Link to conference proceedings)

Publication series

Name	Lecture Notes in Control and Information Sciences
Volume	232
ISSN (Print)	0170-8643
ISSN (Electronic)	1610-7411

Symposium

Symposium	5th International Symposium on Experimental Robotics, ISER 1997
Country/Territory	Spain
City	Barcelona, Catalonia
Period	15/06/97 → 18/06/97
Internet address	https://link.springer.com/book/10.1007/BFb0112945 (Link to conference proceedings)

Access to Document

10.1007/BFb0113002

Cite this

Liu, J., Tang, Y. Y., & Khatib, O. (1997). Modeling and learning robot manipulation strategies. In A. Casals, & A. T. Almeida (Eds.), Experimental Robotics V: The Fifth International Symposium Barcelona, Catalonia, June 15-18, 1997 (1 ed., pp. 687-700). (Lecture Notes in Control and Information Sciences; Vol. 232). Springer Berlin Heidelberg. https://doi.org/10.1007/BFb0113002

@inproceedings{e6f7821f47424481a12c6d9afd59f81c,

title = "Modeling and learning robot manipulation strategies",

abstract = "This paper describes a general approach to learning and planning robot manipulation strategies. Here, the strategies are represented using a discrete-event dynamical systems model where each node corresponds to a state in the robot task environment that triggers certain action schemata and each arc corresponds to a plausible action that brings the task environment into a new state. With such a representation, a manipulation strategy plan can be derived by searching a connected state transition path that is the most reliable. Here, we define the notion of reliability in terms of the estimated chance of success in reaching a desirable state. In the paper, we first present the formalism of discrete-event dynamical system in the context of robot manipulation tasks. Throughout the paper, we provide both illustrative and experimental examples to demonstrate the proposed approach.",

author = "Jiming Liu and Tang, \{Y. Y.\} and Oussama Khatib",

note = "Publisher copyright: {\textcopyright} 1998 Springer-Verlag London Limited; 5th International Symposium on Experimental Robotics, ISER 1997 ; Conference date: 15-06-1997 Through 18-06-1997",

year = "1997",

month = jun,

day = "15",

doi = "10.1007/BFb0113002",

language = "English",

series = "Lecture Notes in Control and Information Sciences",

publisher = "Springer Berlin Heidelberg",

pages = "687--700",

editor = "Alicia Casals and Almeida, \{Anibal T.\}",

booktitle = "Experimental Robotics V",

edition = "1",

url = "https://link.springer.com/book/10.1007/BFb0112945",

}

Liu, J, Tang, YY & Khatib, O 1997, Modeling and learning robot manipulation strategies. in A Casals & AT Almeida (eds), Experimental Robotics V: The Fifth International Symposium Barcelona, Catalonia, June 15-18, 1997. 1 edn, Lecture Notes in Control and Information Sciences, vol. 232, Springer Berlin Heidelberg, pp. 687-700, 5th International Symposium on Experimental Robotics, ISER 1997, Barcelona, Catalonia, Spain, 15/06/97. https://doi.org/10.1007/BFb0113002

Modeling and learning robot manipulation strategies. / Liu, Jiming; Tang, Y. Y.; Khatib, Oussama.
Experimental Robotics V: The Fifth International Symposium Barcelona, Catalonia, June 15-18, 1997. ed. / Alicia Casals; Anibal T. Almeida. 1. ed. Springer Berlin Heidelberg, 1997. p. 687-700 (Lecture Notes in Control and Information Sciences; Vol. 232).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Modeling and learning robot manipulation strategies

AU - Liu, Jiming

AU - Tang, Y. Y.

AU - Khatib, Oussama

PY - 1997/6/15

Y1 - 1997/6/15

N2 - This paper describes a general approach to learning and planning robot manipulation strategies. Here, the strategies are represented using a discrete-event dynamical systems model where each node corresponds to a state in the robot task environment that triggers certain action schemata and each arc corresponds to a plausible action that brings the task environment into a new state. With such a representation, a manipulation strategy plan can be derived by searching a connected state transition path that is the most reliable. Here, we define the notion of reliability in terms of the estimated chance of success in reaching a desirable state. In the paper, we first present the formalism of discrete-event dynamical system in the context of robot manipulation tasks. Throughout the paper, we provide both illustrative and experimental examples to demonstrate the proposed approach.

AB - This paper describes a general approach to learning and planning robot manipulation strategies. Here, the strategies are represented using a discrete-event dynamical systems model where each node corresponds to a state in the robot task environment that triggers certain action schemata and each arc corresponds to a plausible action that brings the task environment into a new state. With such a representation, a manipulation strategy plan can be derived by searching a connected state transition path that is the most reliable. Here, we define the notion of reliability in terms of the estimated chance of success in reaching a desirable state. In the paper, we first present the formalism of discrete-event dynamical system in the context of robot manipulation tasks. Throughout the paper, we provide both illustrative and experimental examples to demonstrate the proposed approach.

UR - https://link.springer.com/book/10.1007/BFb0112945?page=3#toc

U2 - 10.1007/BFb0113002

DO - 10.1007/BFb0113002

M3 - Conference proceeding

T3 - Lecture Notes in Control and Information Sciences

SP - 687

EP - 700

BT - Experimental Robotics V

A2 - Casals, Alicia

A2 - Almeida, Anibal T.

PB - Springer Berlin Heidelberg

T2 - 5th International Symposium on Experimental Robotics, ISER 1997

Y2 - 15 June 1997 through 18 June 1997

ER -

Modeling and learning robot manipulation strategies

Abstract

Publication series

Symposium

Access to Document

Other files and links

Fingerprint

Cite this