An incremental episodic memory framework for topological map building

Wei Hong Chin, Azhar Aulia Saputra, Yuichiro Toda, Naoyuki Kubota

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

Abstract

In this paper, an episodic memory learning framework is proposed for categorizing and encoding sensory information that acquired from a robot for environment adaptation and sensorimotor map building. The proposed learning model termed as Incremental Episodic Memory Adaptive Resonance Theory (In-EMART), consists two layers of ART networks which used to detect novel event encountered by the robot and learn the spatio-temporal relationship by creating neurons incrementally. A set of connected episodes forms a sensorimotor map that can be used for path planning and goal navigation autonomously. The experimental results for a mobile robot show that: (i) In-EMART can learn sensory data in real time which is important for robot implementation; (ii) the model solves the perceptual aliasing issue by recalling the connected episode neurons; (iii) compared with previous works, the proposed method further generates a sensorimotor map for connecting episodes together to navigate from one place to another continuously.

Original languageEnglish
Title of host publicationInternational Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings
EditorsTri Hadiah Muliawati, Muhammad Febrian Ardiansyah, Dewi Mutiara Sari, Desy Intan Permatasari, Mu'arifin
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages322-327
Number of pages6
ISBN (Electronic)9781538680797
DOIs
Publication statusPublished - Jan 28 2019
Externally publishedYes
Event2018 International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Bali, Indonesia
Duration: Oct 29 2018Oct 30 2018

Publication series

NameInternational Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings

Conference

Conference2018 International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018
CountryIndonesia
CityBali
Period10/29/1810/30/18

Fingerprint

Robots
Data storage equipment
Neurons
Motion planning
Mobile robots
Navigation

Keywords

  • Cognitive Map
  • Episodic Memory
  • Robot Navigation
  • SLAM

ASJC Scopus subject areas

  • Artificial Intelligence
  • Signal Processing

Cite this

Chin, W. H., Saputra, A. A., Toda, Y., & Kubota, N. (2019). An incremental episodic memory framework for topological map building. In T. H. Muliawati, M. F. Ardiansyah, D. M. Sari, D. I. Permatasari, & Mu'arifin (Eds.), International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings (pp. 322-327). [8628468] (International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/KCIC.2018.8628468

An incremental episodic memory framework for topological map building. / Chin, Wei Hong; Saputra, Azhar Aulia; Toda, Yuichiro; Kubota, Naoyuki.

International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings. ed. / Tri Hadiah Muliawati; Muhammad Febrian Ardiansyah; Dewi Mutiara Sari; Desy Intan Permatasari; Mu'arifin. Institute of Electrical and Electronics Engineers Inc., 2019. p. 322-327 8628468 (International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings).

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

Chin, WH, Saputra, AA, Toda, Y & Kubota, N 2019, An incremental episodic memory framework for topological map building. in TH Muliawati, MF Ardiansyah, DM Sari, DI Permatasari & Mu'arifin (eds), International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings., 8628468, International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 322-327, 2018 International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018, Bali, Indonesia, 10/29/18. https://doi.org/10.1109/KCIC.2018.8628468
Chin WH, Saputra AA, Toda Y, Kubota N. An incremental episodic memory framework for topological map building. In Muliawati TH, Ardiansyah MF, Sari DM, Permatasari DI, Mu'arifin, editors, International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. p. 322-327. 8628468. (International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings). https://doi.org/10.1109/KCIC.2018.8628468
Chin, Wei Hong ; Saputra, Azhar Aulia ; Toda, Yuichiro ; Kubota, Naoyuki. / An incremental episodic memory framework for topological map building. International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings. editor / Tri Hadiah Muliawati ; Muhammad Febrian Ardiansyah ; Dewi Mutiara Sari ; Desy Intan Permatasari ; Mu'arifin. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 322-327 (International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings).
@inproceedings{fba4f327d6274103b286ef348371eb26,
title = "An incremental episodic memory framework for topological map building",
abstract = "In this paper, an episodic memory learning framework is proposed for categorizing and encoding sensory information that acquired from a robot for environment adaptation and sensorimotor map building. The proposed learning model termed as Incremental Episodic Memory Adaptive Resonance Theory (In-EMART), consists two layers of ART networks which used to detect novel event encountered by the robot and learn the spatio-temporal relationship by creating neurons incrementally. A set of connected episodes forms a sensorimotor map that can be used for path planning and goal navigation autonomously. The experimental results for a mobile robot show that: (i) In-EMART can learn sensory data in real time which is important for robot implementation; (ii) the model solves the perceptual aliasing issue by recalling the connected episode neurons; (iii) compared with previous works, the proposed method further generates a sensorimotor map for connecting episodes together to navigate from one place to another continuously.",
keywords = "Cognitive Map, Episodic Memory, Robot Navigation, SLAM",
author = "Chin, {Wei Hong} and Saputra, {Azhar Aulia} and Yuichiro Toda and Naoyuki Kubota",
year = "2019",
month = "1",
day = "28",
doi = "10.1109/KCIC.2018.8628468",
language = "English",
series = "International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "322--327",
editor = "Muliawati, {Tri Hadiah} and Ardiansyah, {Muhammad Febrian} and Sari, {Dewi Mutiara} and Permatasari, {Desy Intan} and Mu'arifin",
booktitle = "International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings",

}

TY - GEN

T1 - An incremental episodic memory framework for topological map building

AU - Chin, Wei Hong

AU - Saputra, Azhar Aulia

AU - Toda, Yuichiro

AU - Kubota, Naoyuki

PY - 2019/1/28

Y1 - 2019/1/28

N2 - In this paper, an episodic memory learning framework is proposed for categorizing and encoding sensory information that acquired from a robot for environment adaptation and sensorimotor map building. The proposed learning model termed as Incremental Episodic Memory Adaptive Resonance Theory (In-EMART), consists two layers of ART networks which used to detect novel event encountered by the robot and learn the spatio-temporal relationship by creating neurons incrementally. A set of connected episodes forms a sensorimotor map that can be used for path planning and goal navigation autonomously. The experimental results for a mobile robot show that: (i) In-EMART can learn sensory data in real time which is important for robot implementation; (ii) the model solves the perceptual aliasing issue by recalling the connected episode neurons; (iii) compared with previous works, the proposed method further generates a sensorimotor map for connecting episodes together to navigate from one place to another continuously.

AB - In this paper, an episodic memory learning framework is proposed for categorizing and encoding sensory information that acquired from a robot for environment adaptation and sensorimotor map building. The proposed learning model termed as Incremental Episodic Memory Adaptive Resonance Theory (In-EMART), consists two layers of ART networks which used to detect novel event encountered by the robot and learn the spatio-temporal relationship by creating neurons incrementally. A set of connected episodes forms a sensorimotor map that can be used for path planning and goal navigation autonomously. The experimental results for a mobile robot show that: (i) In-EMART can learn sensory data in real time which is important for robot implementation; (ii) the model solves the perceptual aliasing issue by recalling the connected episode neurons; (iii) compared with previous works, the proposed method further generates a sensorimotor map for connecting episodes together to navigate from one place to another continuously.

KW - Cognitive Map

KW - Episodic Memory

KW - Robot Navigation

KW - SLAM

UR - http://www.scopus.com/inward/record.url?scp=85062863614&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062863614&partnerID=8YFLogxK

U2 - 10.1109/KCIC.2018.8628468

DO - 10.1109/KCIC.2018.8628468

M3 - Conference contribution

AN - SCOPUS:85062863614

T3 - International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings

SP - 322

EP - 327

BT - International Electronics Symposium on Knowledge Creation and Intelligent Computing, IES-KCIC 2018 - Proceedings

A2 - Muliawati, Tri Hadiah

A2 - Ardiansyah, Muhammad Febrian

A2 - Sari, Dewi Mutiara

A2 - Permatasari, Desy Intan

A2 - Mu'arifin, null

PB - Institute of Electrical and Electronics Engineers Inc.

ER -