Optimal Policy Generation for Partially Satisfiable Co-Safe LTL Specifications

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Standard

Optimal Policy Generation for Partially Satisfiable Co-Safe LTL Specifications. / Lacerda, B.; Parker, D.; Hawes, N.

Proc. 24th International Joint Conference on Artificial Intelligence (IJCAI'15). ed. / Qiang Yang; Michael Wooldridge. Association for the Advancement of Artificial Intelligence, 2015. p. 1587-1593.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Harvard

Lacerda, B, Parker, D & Hawes, N 2015, Optimal Policy Generation for Partially Satisfiable Co-Safe LTL Specifications. in Q Yang & M Wooldridge (eds), Proc. 24th International Joint Conference on Artificial Intelligence (IJCAI'15). Association for the Advancement of Artificial Intelligence, pp. 1587-1593, International Joint Conference on Artificial Intelligence, 24th (ICJAI 2015), Argentina, 25/07/15. <http://ijcai.org/papers15/Papers/IJCAI15-227.pdf>

APA

Lacerda, B., Parker, D., & Hawes, N. (2015). Optimal Policy Generation for Partially Satisfiable Co-Safe LTL Specifications. In Q. Yang, & M. Wooldridge (Eds.), Proc. 24th International Joint Conference on Artificial Intelligence (IJCAI'15) (pp. 1587-1593). Association for the Advancement of Artificial Intelligence. http://ijcai.org/papers15/Papers/IJCAI15-227.pdf

Vancouver

Lacerda B, Parker D, Hawes N. Optimal Policy Generation for Partially Satisfiable Co-Safe LTL Specifications. In Yang Q, Wooldridge M, editors, Proc. 24th International Joint Conference on Artificial Intelligence (IJCAI'15). Association for the Advancement of Artificial Intelligence. 2015. p. 1587-1593

Author

Lacerda, B. ; Parker, D. ; Hawes, N. / Optimal Policy Generation for Partially Satisfiable Co-Safe LTL Specifications. Proc. 24th International Joint Conference on Artificial Intelligence (IJCAI'15). editor / Qiang Yang ; Michael Wooldridge. Association for the Advancement of Artificial Intelligence, 2015. pp. 1587-1593

Bibtex

@inproceedings{44a94f19fa53403c98aadbecad595077,
title = "Optimal Policy Generation for Partially Satisfiable Co-Safe LTL Specifications",
abstract = "We present a method to calculate cost-optimal policies for co-safe linear temporal logic task specifications over a Markov decision process model of a stochastic system. Our key contribution is to address scenarios in which the task may not be achievable with probability one. We formalise a task progression metric and, using multi-objective probabilistic model checking, generate policies that are formally guaranteed to, in decreasing order of priority: maximise the probability of finishing the task; maximise progress towards completion, if this is not possible; and minimise the expected time or cost required. We illustrate and evaluate our approach in a robot task planning scenario, where the task is to visit a set of rooms that may be inaccessible during execution.",
author = "B. Lacerda and D. Parker and N. Hawes",
year = "2015",
month = jul,
day = "20",
language = "English",
isbn = "9781577357384",
pages = "1587--1593",
editor = "Qiang Yang and Michael Wooldridge",
booktitle = "Proc. 24th International Joint Conference on Artificial Intelligence (IJCAI'15)",
publisher = "Association for the Advancement of Artificial Intelligence",
note = "International Joint Conference on Artificial Intelligence, 24th (ICJAI 2015) ; Conference date: 25-07-2015 Through 31-07-2015",

}

RIS

TY - GEN

T1 - Optimal Policy Generation for Partially Satisfiable Co-Safe LTL Specifications

AU - Lacerda, B.

AU - Parker, D.

AU - Hawes, N.

PY - 2015/7/20

Y1 - 2015/7/20

N2 - We present a method to calculate cost-optimal policies for co-safe linear temporal logic task specifications over a Markov decision process model of a stochastic system. Our key contribution is to address scenarios in which the task may not be achievable with probability one. We formalise a task progression metric and, using multi-objective probabilistic model checking, generate policies that are formally guaranteed to, in decreasing order of priority: maximise the probability of finishing the task; maximise progress towards completion, if this is not possible; and minimise the expected time or cost required. We illustrate and evaluate our approach in a robot task planning scenario, where the task is to visit a set of rooms that may be inaccessible during execution.

AB - We present a method to calculate cost-optimal policies for co-safe linear temporal logic task specifications over a Markov decision process model of a stochastic system. Our key contribution is to address scenarios in which the task may not be achievable with probability one. We formalise a task progression metric and, using multi-objective probabilistic model checking, generate policies that are formally guaranteed to, in decreasing order of priority: maximise the probability of finishing the task; maximise progress towards completion, if this is not possible; and minimise the expected time or cost required. We illustrate and evaluate our approach in a robot task planning scenario, where the task is to visit a set of rooms that may be inaccessible during execution.

M3 - Conference contribution

SN - 9781577357384

SP - 1587

EP - 1593

BT - Proc. 24th International Joint Conference on Artificial Intelligence (IJCAI'15)

A2 - Yang, Qiang

A2 - Wooldridge, Michael

PB - Association for the Advancement of Artificial Intelligence

T2 - International Joint Conference on Artificial Intelligence, 24th (ICJAI 2015)

Y2 - 25 July 2015 through 31 July 2015

ER -