Episodic memory multimodal learning for robot sensorimotor map building and navigation

Wei Hong Chin, Yuichiro Toda, Naoyuki Kubota, Chu Kiong Loo, Manjeevan Seera

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

In this paper, an unsupervised learning model of episodic memory is proposed. The proposed model, Enhanced Episodic Memory Adaptive Resonance Theory (EEM-ART), categorizes and encodes experiences of a robot to the environment and generates a cognitive map. EEM-ART consists of multi-layer ART networks to extract novel events and encode spatio-temporal connection as episodes by incrementally generating cognitive neurons. The model connects episodes to construct a sensorimotor map for the robot to continuously perform path planning and goal navigation. Experimental results for a mobile robot indicate that EEM-ART can process multiple sensory sources for learning events and encoding episodes simultaneously. The model overcomes perceptual aliasing and robot localization by recalling the encoded episodes with a new anticipation function and generates sensorimotor map to connect episodes together to execute tasks continuously with little to no human intervention.

Original languageEnglish
JournalIEEE Transactions on Cognitive and Developmental Systems
DOIs
Publication statusAccepted/In press - Jan 1 2018
Externally publishedYes

Fingerprint

Navigation
Robots
Data storage equipment
Unsupervised learning
Network layers
Motion planning
Mobile robots
Neurons

Keywords

  • Adaptive Resonance Theory
  • Episodic Memory
  • Navigation
  • Neurons
  • Path planning
  • Robot Navigation.
  • Robot sensing systems
  • Subspace constraints
  • Task analysis

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

Cite this

Episodic memory multimodal learning for robot sensorimotor map building and navigation. / Chin, Wei Hong; Toda, Yuichiro; Kubota, Naoyuki; Loo, Chu Kiong; Seera, Manjeevan.

In: IEEE Transactions on Cognitive and Developmental Systems, 01.01.2018.

Research output: Contribution to journalArticle

@article{6296ae07a09f43459537f55f49d7038f,
title = "Episodic memory multimodal learning for robot sensorimotor map building and navigation",
abstract = "In this paper, an unsupervised learning model of episodic memory is proposed. The proposed model, Enhanced Episodic Memory Adaptive Resonance Theory (EEM-ART), categorizes and encodes experiences of a robot to the environment and generates a cognitive map. EEM-ART consists of multi-layer ART networks to extract novel events and encode spatio-temporal connection as episodes by incrementally generating cognitive neurons. The model connects episodes to construct a sensorimotor map for the robot to continuously perform path planning and goal navigation. Experimental results for a mobile robot indicate that EEM-ART can process multiple sensory sources for learning events and encoding episodes simultaneously. The model overcomes perceptual aliasing and robot localization by recalling the encoded episodes with a new anticipation function and generates sensorimotor map to connect episodes together to execute tasks continuously with little to no human intervention.",
keywords = "Adaptive Resonance Theory, Episodic Memory, Navigation, Neurons, Path planning, Robot Navigation., Robot sensing systems, Subspace constraints, Task analysis",
author = "Chin, {Wei Hong} and Yuichiro Toda and Naoyuki Kubota and Loo, {Chu Kiong} and Manjeevan Seera",
year = "2018",
month = "1",
day = "1",
doi = "10.1109/TCDS.2018.2875309",
language = "English",
journal = "IEEE Transactions on Cognitive and Developmental Systems",
issn = "2379-8920",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Episodic memory multimodal learning for robot sensorimotor map building and navigation

AU - Chin, Wei Hong

AU - Toda, Yuichiro

AU - Kubota, Naoyuki

AU - Loo, Chu Kiong

AU - Seera, Manjeevan

PY - 2018/1/1

Y1 - 2018/1/1

N2 - In this paper, an unsupervised learning model of episodic memory is proposed. The proposed model, Enhanced Episodic Memory Adaptive Resonance Theory (EEM-ART), categorizes and encodes experiences of a robot to the environment and generates a cognitive map. EEM-ART consists of multi-layer ART networks to extract novel events and encode spatio-temporal connection as episodes by incrementally generating cognitive neurons. The model connects episodes to construct a sensorimotor map for the robot to continuously perform path planning and goal navigation. Experimental results for a mobile robot indicate that EEM-ART can process multiple sensory sources for learning events and encoding episodes simultaneously. The model overcomes perceptual aliasing and robot localization by recalling the encoded episodes with a new anticipation function and generates sensorimotor map to connect episodes together to execute tasks continuously with little to no human intervention.

AB - In this paper, an unsupervised learning model of episodic memory is proposed. The proposed model, Enhanced Episodic Memory Adaptive Resonance Theory (EEM-ART), categorizes and encodes experiences of a robot to the environment and generates a cognitive map. EEM-ART consists of multi-layer ART networks to extract novel events and encode spatio-temporal connection as episodes by incrementally generating cognitive neurons. The model connects episodes to construct a sensorimotor map for the robot to continuously perform path planning and goal navigation. Experimental results for a mobile robot indicate that EEM-ART can process multiple sensory sources for learning events and encoding episodes simultaneously. The model overcomes perceptual aliasing and robot localization by recalling the encoded episodes with a new anticipation function and generates sensorimotor map to connect episodes together to execute tasks continuously with little to no human intervention.

KW - Adaptive Resonance Theory

KW - Episodic Memory

KW - Navigation

KW - Neurons

KW - Path planning

KW - Robot Navigation.

KW - Robot sensing systems

KW - Subspace constraints

KW - Task analysis

UR - http://www.scopus.com/inward/record.url?scp=85054637979&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054637979&partnerID=8YFLogxK

U2 - 10.1109/TCDS.2018.2875309

DO - 10.1109/TCDS.2018.2875309

M3 - Article

AN - SCOPUS:85054637979

JO - IEEE Transactions on Cognitive and Developmental Systems

JF - IEEE Transactions on Cognitive and Developmental Systems

SN - 2379-8920

ER -