by Michael Beetz
Abstract:
Autonomous robots that perform complex jobs in changing environments must be capable of managing their plans as the environmental conditions or their tasks change. This raises the problem of deciding whether, when, where, and how to revise the plans as the robots' beliefs change. This article investigates an approach to execution time plan management in which the plans themselves specify the plan adaptation processes. In this approach the robot makes strategical (farsighted) adaptations while it executes a plan using tactical (immediate) decisions and overwrites tactical adaptations after strategical decisions have been reached (if necessary). We present experiments in which the plan adaptation technique is used for the control of two autonomous mobile robots. In one of them it controlled the course of action of a museums tourguide robot that has operated for thirteen days and performed about 3200 plan adaptations reliably.
Reference:
Michael Beetz, "Plan Management for Robotic Agents", In KI - Künstliche Intelligenz; Special Issue on Planning and Scheduling, vol. 15, no. 2, pp. 12–17, 2001.
Bibtex Entry:
@Article{Bee01Pla,
author = "Michael Beetz",
title = "Plan Management for Robotic Agents",
journal = {KI - K{\"u}nstliche Intelligenz; Special Issue on Planning and Scheduling},
volume = {15},
number = {2},
year = "2001",
pages = {12--17},
bib2html_pubtype = {Journal},
bib2html_rescat = {Plan-based Robot Control},
bib2html_groups = {IAS},
bib2html_funding = {ignore},
bib2html_keywords = {Robot, Planning},
abstract = {Autonomous robots that perform complex jobs in changing environments must be capable of managing
their plans as the environmental conditions or their tasks change. This raises the problem of
deciding whether, when, where, and how to revise the plans as the robots' beliefs change. This
article investigates an approach to execution time plan management in which the plans themselves
specify the plan adaptation processes. In this approach the robot makes strategical (farsighted)
adaptations while it executes a plan using tactical (immediate) decisions and overwrites tactical
adaptations after strategical decisions have been reached (if necessary). We present experiments in
which the plan adaptation technique is used for the control of two autonomous mobile robots. In one
of them it controlled the course of action of a museums tourguide robot that has operated for
thirteen days and performed about 3200 plan adaptations reliably.}
}