Using learned action models in execution monitoring

M. Fox, J. Gough, D. Long, R. Qu (Editor)

Research output: Contribution to conferencePaper

Abstract

Planners reason with abstracted models of the behaviours they use to construct plans. When plans are turned into the instructions that drive an executive, the real behaviours interacting with the unpredictable uncertainties of the environment can lead to failure. One of the challenges for intelligent autonomy is to recognise when the actual execution of a behaviour has diverged so far from the expected behaviour that it can be considered to be a failure. In this paper we present further developments of the work described in (Fox et al. 2006), where models of behaviours were learned as Hidden Markov Models. Execution of behaviours is monitored by tracking the most likely trajectory through such a learned model, while possible failures in execution are identified as deviations from common patterns of trajectories within the learned models. We present results for our experiments with a model learned for a robot behaviour.

Conference

Conference25th Workshop of the UK Planning and Scheduling Special Interest Group
Abbreviated titlePlanSIG 2006
CountryUnited Kingdom
CityNottingham
Period14/12/0615/12/06

Fingerprint

Monitoring
Trajectories
Hidden Markov models
Robots
Experiments
Uncertainty

Keywords

  • planning
  • execution monitoring
  • learned action models
  • intelligent autonomy
  • Markov models

Cite this

Fox, M., Gough, J., Long, D., & Qu, R. (Ed.) (2006). Using learned action models in execution monitoring. Paper presented at 25th Workshop of the UK Planning and Scheduling Special Interest Group, Nottingham, United Kingdom.
Fox, M. ; Gough, J. ; Long, D. ; Qu, R. (Editor). / Using learned action models in execution monitoring. Paper presented at 25th Workshop of the UK Planning and Scheduling Special Interest Group, Nottingham, United Kingdom.8 p.
@conference{9cc3033cd70f44de9909a6019039217a,
title = "Using learned action models in execution monitoring",
abstract = "Planners reason with abstracted models of the behaviours they use to construct plans. When plans are turned into the instructions that drive an executive, the real behaviours interacting with the unpredictable uncertainties of the environment can lead to failure. One of the challenges for intelligent autonomy is to recognise when the actual execution of a behaviour has diverged so far from the expected behaviour that it can be considered to be a failure. In this paper we present further developments of the work described in (Fox et al. 2006), where models of behaviours were learned as Hidden Markov Models. Execution of behaviours is monitored by tracking the most likely trajectory through such a learned model, while possible failures in execution are identified as deviations from common patterns of trajectories within the learned models. We present results for our experiments with a model learned for a robot behaviour.",
keywords = "planning, execution monitoring, learned action models, intelligent autonomy, Markov models",
author = "M. Fox and J. Gough and D. Long and R. Qu",
year = "2006",
language = "English",
note = "25th Workshop of the UK Planning and Scheduling Special Interest Group, PlanSIG 2006 ; Conference date: 14-12-2006 Through 15-12-2006",

}

Fox, M, Gough, J, Long, D & Qu, R (ed.) 2006, 'Using learned action models in execution monitoring' Paper presented at 25th Workshop of the UK Planning and Scheduling Special Interest Group, Nottingham, United Kingdom, 14/12/06 - 15/12/06, .

Using learned action models in execution monitoring. / Fox, M.; Gough, J.; Long, D.; Qu, R. (Editor).

2006. Paper presented at 25th Workshop of the UK Planning and Scheduling Special Interest Group, Nottingham, United Kingdom.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Using learned action models in execution monitoring

AU - Fox, M.

AU - Gough, J.

AU - Long, D.

A2 - Qu, R.

PY - 2006

Y1 - 2006

N2 - Planners reason with abstracted models of the behaviours they use to construct plans. When plans are turned into the instructions that drive an executive, the real behaviours interacting with the unpredictable uncertainties of the environment can lead to failure. One of the challenges for intelligent autonomy is to recognise when the actual execution of a behaviour has diverged so far from the expected behaviour that it can be considered to be a failure. In this paper we present further developments of the work described in (Fox et al. 2006), where models of behaviours were learned as Hidden Markov Models. Execution of behaviours is monitored by tracking the most likely trajectory through such a learned model, while possible failures in execution are identified as deviations from common patterns of trajectories within the learned models. We present results for our experiments with a model learned for a robot behaviour.

AB - Planners reason with abstracted models of the behaviours they use to construct plans. When plans are turned into the instructions that drive an executive, the real behaviours interacting with the unpredictable uncertainties of the environment can lead to failure. One of the challenges for intelligent autonomy is to recognise when the actual execution of a behaviour has diverged so far from the expected behaviour that it can be considered to be a failure. In this paper we present further developments of the work described in (Fox et al. 2006), where models of behaviours were learned as Hidden Markov Models. Execution of behaviours is monitored by tracking the most likely trajectory through such a learned model, while possible failures in execution are identified as deviations from common patterns of trajectories within the learned models. We present results for our experiments with a model learned for a robot behaviour.

KW - planning

KW - execution monitoring

KW - learned action models

KW - intelligent autonomy

KW - Markov models

UR - http://www.cis.strath.ac.uk/research/publications/papers/strath_cis_publication_1812.pdf

M3 - Paper

ER -

Fox M, Gough J, Long D, Qu R, (ed.). Using learned action models in execution monitoring. 2006. Paper presented at 25th Workshop of the UK Planning and Scheduling Special Interest Group, Nottingham, United Kingdom.