Towards an architecture for knowledge representation and reasoning in robotics

Shiqi Zhang; Mohan Sridharan; Michael Gelfond; Jeremy Wyatt

doi:10.1007/978-3-319-11973-1_41

Towards an architecture for knowledge representation and reasoning in robotics

Shiqi Zhang, Mohan Sridharan, Michael Gelfond, Jeremy Wyatt

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

Abstract

This paper describes an architecture that combines the complementary strengths of probabilistic graphical models and declarative programming to enable robots to represent and reason with qualitative and quantitative descriptions of uncertainty and domain knowledge. An action language is used for the architecture’s low-level (LL) and high-level (HL) system descriptions, and the HL definition of recorded history is expanded to allow prioritized defaults. For any given objective, tentative plans created in the HL using commonsense reasoning are implemented in the LL using probabilistic algorithms, and the corresponding observations are added to the HL history. Tight coupling between the levels helps automate the selection of relevant variables and the generation of policies in the LL for each HL action, and supports reasoning with violation of defaults, noisy observations and unreliable actions in complex domains. The architecture is evaluated in simulation and on robots moving objects in indoor domains.

Original language	English
Title of host publication	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Publisher	Springer
Pages	400-410
Number of pages	11
Volume	8755
ISBN (Print)	9783319119724
DOIs	https://doi.org/10.1007/978-3-319-11973-1_41
Publication status	Published - 24 Sept 2014
Event	Sixth International Conference on Social Robotics - Sydney, Australia Duration: 27 Oct 2014 → 29 Oct 2014 Conference number: 6

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Publisher	Springer
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	Sixth International Conference on Social Robotics
Abbreviated title	ICSR 2014
Country/Territory	Australia
City	Sydney
Period	27/10/14 → 29/10/14

Access to Document

10.1007/978-3-319-11973-1_41Licence: All Rights Reserved

https://link.springer.com/chapter/10.1007/978-3-319-11973-1_41#Abs1Licence: All Rights Reserved

Cite this

Zhang, S., Sridharan, M., Gelfond, M., & Wyatt, J. (2014). Towards an architecture for knowledge representation and reasoning in robotics. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8755, pp. 400-410). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). Springer. https://doi.org/10.1007/978-3-319-11973-1_41

Zhang, Shiqi ; Sridharan, Mohan ; Gelfond, Michael et al. / Towards an architecture for knowledge representation and reasoning in robotics. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8755 Springer, 2014. pp. 400-410 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{6ec9d6fa67184f2c978b4afabb0c7a32,

title = "Towards an architecture for knowledge representation and reasoning in robotics",

abstract = "This paper describes an architecture that combines the complementary strengths of probabilistic graphical models and declarative programming to enable robots to represent and reason with qualitative and quantitative descriptions of uncertainty and domain knowledge. An action language is used for the architecture{\textquoteright}s low-level (LL) and high-level (HL) system descriptions, and the HL definition of recorded history is expanded to allow prioritized defaults. For any given objective, tentative plans created in the HL using commonsense reasoning are implemented in the LL using probabilistic algorithms, and the corresponding observations are added to the HL history. Tight coupling between the levels helps automate the selection of relevant variables and the generation of policies in the LL for each HL action, and supports reasoning with violation of defaults, noisy observations and unreliable actions in complex domains. The architecture is evaluated in simulation and on robots moving objects in indoor domains.",

author = "Shiqi Zhang and Mohan Sridharan and Michael Gelfond and Jeremy Wyatt",

note = "Sixth International Conference on Social Robotics (ICSR 2014) ; Conference date: 27-10-2014 Through 29-10-2014; Sixth International Conference on Social Robotics, ICSR 2014 ; Conference date: 27-10-2014 Through 29-10-2014",

year = "2014",

month = sep,

day = "24",

doi = "10.1007/978-3-319-11973-1_41",

language = "English",

isbn = "9783319119724",

volume = "8755",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "400--410",

booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

address = "United Kingdom",

}

Zhang, S, Sridharan, M, Gelfond, M & Wyatt, J 2014, Towards an architecture for knowledge representation and reasoning in robotics. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 8755, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Springer, pp. 400-410, Sixth International Conference on Social Robotics, Sydney, Australia, 27/10/14. https://doi.org/10.1007/978-3-319-11973-1_41

Towards an architecture for knowledge representation and reasoning in robotics. / Zhang, Shiqi; Sridharan, Mohan; Gelfond, Michael et al.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8755 Springer, 2014. p. 400-410 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

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

TY - GEN

T1 - Towards an architecture for knowledge representation and reasoning in robotics

AU - Zhang, Shiqi

AU - Sridharan, Mohan

AU - Gelfond, Michael

AU - Wyatt, Jeremy

N1 - Conference code: 6

PY - 2014/9/24

Y1 - 2014/9/24

N2 - This paper describes an architecture that combines the complementary strengths of probabilistic graphical models and declarative programming to enable robots to represent and reason with qualitative and quantitative descriptions of uncertainty and domain knowledge. An action language is used for the architecture’s low-level (LL) and high-level (HL) system descriptions, and the HL definition of recorded history is expanded to allow prioritized defaults. For any given objective, tentative plans created in the HL using commonsense reasoning are implemented in the LL using probabilistic algorithms, and the corresponding observations are added to the HL history. Tight coupling between the levels helps automate the selection of relevant variables and the generation of policies in the LL for each HL action, and supports reasoning with violation of defaults, noisy observations and unreliable actions in complex domains. The architecture is evaluated in simulation and on robots moving objects in indoor domains.

AB - This paper describes an architecture that combines the complementary strengths of probabilistic graphical models and declarative programming to enable robots to represent and reason with qualitative and quantitative descriptions of uncertainty and domain knowledge. An action language is used for the architecture’s low-level (LL) and high-level (HL) system descriptions, and the HL definition of recorded history is expanded to allow prioritized defaults. For any given objective, tentative plans created in the HL using commonsense reasoning are implemented in the LL using probabilistic algorithms, and the corresponding observations are added to the HL history. Tight coupling between the levels helps automate the selection of relevant variables and the generation of policies in the LL for each HL action, and supports reasoning with violation of defaults, noisy observations and unreliable actions in complex domains. The architecture is evaluated in simulation and on robots moving objects in indoor domains.

U2 - 10.1007/978-3-319-11973-1_41

DO - 10.1007/978-3-319-11973-1_41

M3 - Conference contribution

SN - 9783319119724

VL - 8755

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 400

EP - 410

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer

T2 - Sixth International Conference on Social Robotics

Y2 - 27 October 2014 through 29 October 2014

ER -

Zhang S, Sridharan M, Gelfond M, Wyatt J. Towards an architecture for knowledge representation and reasoning in robotics. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8755. Springer. 2014. p. 400-410. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-11973-1_41

Towards an architecture for knowledge representation and reasoning in robotics

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this