Towards autonomously predicting and learning a Robot's efficiency in performing tasks

Ayan Dutta, Prithviraj Dasgupta, José Baca, Carl A Nelson

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

Abstract

We consider the problem of predicting and learning the efficiency of navigation tasks being performed by a robot in an initially unknown or partially known, unstructured environment. Predicting the efficiency of a task is a crucial problem in unstructured environments as it enables the robot(s) to decide whether it is feasible to perform the task under the current environment conditions and within its available resources such as battery power. This problem is non-trivial as robot's performance changes during operation based on its operational conditions, available battery change, etc. In this paper, we have addressed this problem by using a learning-based technique that the robot uses to predict its expected efficiency for performing a new task based on the task's similarity and recentness with previously performed tasks. Experimental results show that our proposed technique can successfully predict the efficiency of a task from previous task experiences and this prediction gets better with number of tasks performed. We have also shown empirically that our model is robust to changes in environmental conditions such as localization and wheel slip noise1.

Original languageEnglish (US)
Title of host publicationProceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013
Pages92-95
Number of pages4
DOIs
StatePublished - Dec 1 2013
Event2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology Workshops, WI-IATW 2013 - Atlanta, GA, United States
Duration: Nov 17 2013Nov 20 2013

Publication series

NameProceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013
Volume3

Conference

Conference2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology Workshops, WI-IATW 2013
CountryUnited States
CityAtlanta, GA
Period11/17/1311/20/13

Fingerprint

Robots
Wheels
Navigation

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Dutta, A., Dasgupta, P., Baca, J., & Nelson, C. A. (2013). Towards autonomously predicting and learning a Robot's efficiency in performing tasks. In Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013 (pp. 92-95). [6690702] (Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013; Vol. 3). https://doi.org/10.1109/WI-IAT.2013.157

Towards autonomously predicting and learning a Robot's efficiency in performing tasks. / Dutta, Ayan; Dasgupta, Prithviraj; Baca, José; Nelson, Carl A.

Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013. 2013. p. 92-95 6690702 (Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013; Vol. 3).

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

Dutta, A, Dasgupta, P, Baca, J & Nelson, CA 2013, Towards autonomously predicting and learning a Robot's efficiency in performing tasks. in Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013., 6690702, Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013, vol. 3, pp. 92-95, 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology Workshops, WI-IATW 2013, Atlanta, GA, United States, 11/17/13. https://doi.org/10.1109/WI-IAT.2013.157
Dutta A, Dasgupta P, Baca J, Nelson CA. Towards autonomously predicting and learning a Robot's efficiency in performing tasks. In Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013. 2013. p. 92-95. 6690702. (Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013). https://doi.org/10.1109/WI-IAT.2013.157
Dutta, Ayan ; Dasgupta, Prithviraj ; Baca, José ; Nelson, Carl A. / Towards autonomously predicting and learning a Robot's efficiency in performing tasks. Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013. 2013. pp. 92-95 (Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013).
@inproceedings{840efe0032e44d6d8a46e6a613c5207d,
title = "Towards autonomously predicting and learning a Robot's efficiency in performing tasks",
abstract = "We consider the problem of predicting and learning the efficiency of navigation tasks being performed by a robot in an initially unknown or partially known, unstructured environment. Predicting the efficiency of a task is a crucial problem in unstructured environments as it enables the robot(s) to decide whether it is feasible to perform the task under the current environment conditions and within its available resources such as battery power. This problem is non-trivial as robot's performance changes during operation based on its operational conditions, available battery change, etc. In this paper, we have addressed this problem by using a learning-based technique that the robot uses to predict its expected efficiency for performing a new task based on the task's similarity and recentness with previously performed tasks. Experimental results show that our proposed technique can successfully predict the efficiency of a task from previous task experiences and this prediction gets better with number of tasks performed. We have also shown empirically that our model is robust to changes in environmental conditions such as localization and wheel slip noise1.",
author = "Ayan Dutta and Prithviraj Dasgupta and Jos{\'e} Baca and Nelson, {Carl A}",
year = "2013",
month = "12",
day = "1",
doi = "10.1109/WI-IAT.2013.157",
language = "English (US)",
isbn = "9781479929023",
series = "Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013",
pages = "92--95",
booktitle = "Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013",

}

TY - GEN

T1 - Towards autonomously predicting and learning a Robot's efficiency in performing tasks

AU - Dutta, Ayan

AU - Dasgupta, Prithviraj

AU - Baca, José

AU - Nelson, Carl A

PY - 2013/12/1

Y1 - 2013/12/1

N2 - We consider the problem of predicting and learning the efficiency of navigation tasks being performed by a robot in an initially unknown or partially known, unstructured environment. Predicting the efficiency of a task is a crucial problem in unstructured environments as it enables the robot(s) to decide whether it is feasible to perform the task under the current environment conditions and within its available resources such as battery power. This problem is non-trivial as robot's performance changes during operation based on its operational conditions, available battery change, etc. In this paper, we have addressed this problem by using a learning-based technique that the robot uses to predict its expected efficiency for performing a new task based on the task's similarity and recentness with previously performed tasks. Experimental results show that our proposed technique can successfully predict the efficiency of a task from previous task experiences and this prediction gets better with number of tasks performed. We have also shown empirically that our model is robust to changes in environmental conditions such as localization and wheel slip noise1.

AB - We consider the problem of predicting and learning the efficiency of navigation tasks being performed by a robot in an initially unknown or partially known, unstructured environment. Predicting the efficiency of a task is a crucial problem in unstructured environments as it enables the robot(s) to decide whether it is feasible to perform the task under the current environment conditions and within its available resources such as battery power. This problem is non-trivial as robot's performance changes during operation based on its operational conditions, available battery change, etc. In this paper, we have addressed this problem by using a learning-based technique that the robot uses to predict its expected efficiency for performing a new task based on the task's similarity and recentness with previously performed tasks. Experimental results show that our proposed technique can successfully predict the efficiency of a task from previous task experiences and this prediction gets better with number of tasks performed. We have also shown empirically that our model is robust to changes in environmental conditions such as localization and wheel slip noise1.

UR - http://www.scopus.com/inward/record.url?scp=84893258907&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893258907&partnerID=8YFLogxK

U2 - 10.1109/WI-IAT.2013.157

DO - 10.1109/WI-IAT.2013.157

M3 - Conference contribution

SN - 9781479929023

T3 - Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013

SP - 92

EP - 95

BT - Proceedings - 2013 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IATW 2013

ER -