A temporal logic programming approach to planning

Kai Yang; Cong Tian; Nan Zhang; Zhenhua Duan; Hongwei Du

doi:10.1007/s10878-019-00389-y

A temporal logic programming approach to planning

Kai Yang
, Cong Tian^*
, Nan Zhang
, Zhenhua Duan
, Hongwei Du

^*Corresponding author for this work

State Key Laboratory of Integrated Services Networks
Harbin Institute of Technology Shenzhen

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

This paper presents an approach to performing artificial intelligence planning through temporal logic programming with Search Control Knowledge (SCK). First, the planning problem described with Planning Domain Description Language is modeled as a program m in Modeling, Simulation and Verification Language (MSVL). Second, the SCK is also specified with an MSVL program m^′. Third, using the basic operation “and” in MSVL, a new MSVL program “mandm′” is obtained. Forth, with the compiler MC of MSVL, an executable binary code of program “mandm′” is obtained. Finally, planning result can be obtained via executing the executable code. Experimental results on selected benchmark planning domains from the International Planning Competition 2014 show that our approach is more effective in practice. Furthermore, the obtained plans are verified with the toolkit MSV so that a plan can be confirmed whether it is a reliable one.

Original language	English
Pages (from-to)	402-420
Number of pages	19
Journal	Journal of Combinatorial Optimization
Volume	38
Issue number	2
DOIs	https://doi.org/10.1007/s10878-019-00389-y
State	Published - 15 Aug 2019
Externally published	Yes

Keywords

MSVL
PDDL
Planning
Search control knowledge
Temporal logic

Access to Document

10.1007/s10878-019-00389-y

Cite this

@article{6cfed52ba2d7486da837c86c2ae70be7,

title = "A temporal logic programming approach to planning",

abstract = "This paper presents an approach to performing artificial intelligence planning through temporal logic programming with Search Control Knowledge (SCK). First, the planning problem described with Planning Domain Description Language is modeled as a program m in Modeling, Simulation and Verification Language (MSVL). Second, the SCK is also specified with an MSVL program m′. Third, using the basic operation “and” in MSVL, a new MSVL program “mandm′” is obtained. Forth, with the compiler MC of MSVL, an executable binary code of program “mandm′” is obtained. Finally, planning result can be obtained via executing the executable code. Experimental results on selected benchmark planning domains from the International Planning Competition 2014 show that our approach is more effective in practice. Furthermore, the obtained plans are verified with the toolkit MSV so that a plan can be confirmed whether it is a reliable one.",

keywords = "MSVL, PDDL, Planning, Search control knowledge, Temporal logic",

author = "Kai Yang and Cong Tian and Nan Zhang and Zhenhua Duan and Hongwei Du",

note = "Publisher Copyright: {\textcopyright} 2019, Springer Science+Business Media, LLC, part of Springer Nature.",

year = "2019",

month = aug,

day = "15",

doi = "10.1007/s10878-019-00389-y",

language = "英语",

volume = "38",

pages = "402--420",

journal = "Journal of Combinatorial Optimization",

issn = "1382-6905",

publisher = "Springer Netherlands",

number = "2",

}

TY - JOUR

T1 - A temporal logic programming approach to planning

AU - Yang, Kai

AU - Tian, Cong

AU - Zhang, Nan

AU - Duan, Zhenhua

AU - Du, Hongwei

PY - 2019/8/15

Y1 - 2019/8/15

N2 - This paper presents an approach to performing artificial intelligence planning through temporal logic programming with Search Control Knowledge (SCK). First, the planning problem described with Planning Domain Description Language is modeled as a program m in Modeling, Simulation and Verification Language (MSVL). Second, the SCK is also specified with an MSVL program m′. Third, using the basic operation “and” in MSVL, a new MSVL program “mandm′” is obtained. Forth, with the compiler MC of MSVL, an executable binary code of program “mandm′” is obtained. Finally, planning result can be obtained via executing the executable code. Experimental results on selected benchmark planning domains from the International Planning Competition 2014 show that our approach is more effective in practice. Furthermore, the obtained plans are verified with the toolkit MSV so that a plan can be confirmed whether it is a reliable one.

AB - This paper presents an approach to performing artificial intelligence planning through temporal logic programming with Search Control Knowledge (SCK). First, the planning problem described with Planning Domain Description Language is modeled as a program m in Modeling, Simulation and Verification Language (MSVL). Second, the SCK is also specified with an MSVL program m′. Third, using the basic operation “and” in MSVL, a new MSVL program “mandm′” is obtained. Forth, with the compiler MC of MSVL, an executable binary code of program “mandm′” is obtained. Finally, planning result can be obtained via executing the executable code. Experimental results on selected benchmark planning domains from the International Planning Competition 2014 show that our approach is more effective in practice. Furthermore, the obtained plans are verified with the toolkit MSV so that a plan can be confirmed whether it is a reliable one.

KW - MSVL

KW - PDDL

KW - Planning

KW - Search control knowledge

KW - Temporal logic

UR - https://www.scopus.com/pages/publications/85061335960

U2 - 10.1007/s10878-019-00389-y

DO - 10.1007/s10878-019-00389-y

M3 - 文章

AN - SCOPUS:85061335960

SN - 1382-6905

VL - 38

SP - 402

EP - 420

JO - Journal of Combinatorial Optimization

JF - Journal of Combinatorial Optimization

IS - 2

ER -

A temporal logic programming approach to planning

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this