Ensemble of Bayesian filter with active and passive nodes for loop closure detection

Mohammed Omar Salameh, Azizi Abdullah, Shahnorbanun Saran

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

Abstract

In this paper, we describe a novel extension of the real-time appearance-based mapping (RTAB-Map), called the Ensemble of Real-Time Appearance-Based Mapping (ERTAB-Map). The original RTAB-Map calculates the probabilities of multiple beliefs for loop closure detection based on a single descriptor model. However, the ERTAB-Map can use an arbitrary number of descriptor models, in which a set of probability belief models are evaluated using an ensemble learning approach. The probability values are extracted from the active working memory and the passive long term memory of RTAB-Map. We have performed experiments on 388 images from the Lib6Indoor and 1063 images from Lib6Outdoor datasets. The results show that our ensemble of active and passive outperforms the original RTAB-Map. Furthermore, the ensemble achieves a recall of 91.59% and 98.65% on the Lib6Indoor and Lib6Outdoor respectively, with a corresponding precision of 100%.

Original languageEnglish
Title of host publication2017 18th International Conference on Advanced Robotics, ICAR 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages482-486
Number of pages5
ISBN (Electronic)9781538631577
DOIs
Publication statusPublished - 30 Aug 2017
Event18th International Conference on Advanced Robotics, ICAR 2017 - Hong Kong, China
Duration: 10 Jul 201712 Jul 2017

Other

Other18th International Conference on Advanced Robotics, ICAR 2017
CountryChina
CityHong Kong
Period10/7/1712/7/17

Fingerprint

Closure
Ensemble
Filter
Real-time
Vertex of a graph
Descriptors
Data storage equipment
Working Memory
Ensemble Learning
Memory Term
Model
Calculate
Arbitrary
Experiment
Experiments
Beliefs

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Mechanical Engineering
  • Control and Optimization

Cite this

Salameh, M. O., Abdullah, A., & Saran, S. (2017). Ensemble of Bayesian filter with active and passive nodes for loop closure detection. In 2017 18th International Conference on Advanced Robotics, ICAR 2017 (pp. 482-486). [8023653] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICAR.2017.8023653

Ensemble of Bayesian filter with active and passive nodes for loop closure detection. / Salameh, Mohammed Omar; Abdullah, Azizi; Saran, Shahnorbanun.

2017 18th International Conference on Advanced Robotics, ICAR 2017. Institute of Electrical and Electronics Engineers Inc., 2017. p. 482-486 8023653.

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

Salameh, MO, Abdullah, A & Saran, S 2017, Ensemble of Bayesian filter with active and passive nodes for loop closure detection. in 2017 18th International Conference on Advanced Robotics, ICAR 2017., 8023653, Institute of Electrical and Electronics Engineers Inc., pp. 482-486, 18th International Conference on Advanced Robotics, ICAR 2017, Hong Kong, China, 10/7/17. https://doi.org/10.1109/ICAR.2017.8023653
Salameh MO, Abdullah A, Saran S. Ensemble of Bayesian filter with active and passive nodes for loop closure detection. In 2017 18th International Conference on Advanced Robotics, ICAR 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 482-486. 8023653 https://doi.org/10.1109/ICAR.2017.8023653
Salameh, Mohammed Omar ; Abdullah, Azizi ; Saran, Shahnorbanun. / Ensemble of Bayesian filter with active and passive nodes for loop closure detection. 2017 18th International Conference on Advanced Robotics, ICAR 2017. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 482-486
@inproceedings{061a576108f24bfc850e2e76e7b4e7ef,
title = "Ensemble of Bayesian filter with active and passive nodes for loop closure detection",
abstract = "In this paper, we describe a novel extension of the real-time appearance-based mapping (RTAB-Map), called the Ensemble of Real-Time Appearance-Based Mapping (ERTAB-Map). The original RTAB-Map calculates the probabilities of multiple beliefs for loop closure detection based on a single descriptor model. However, the ERTAB-Map can use an arbitrary number of descriptor models, in which a set of probability belief models are evaluated using an ensemble learning approach. The probability values are extracted from the active working memory and the passive long term memory of RTAB-Map. We have performed experiments on 388 images from the Lib6Indoor and 1063 images from Lib6Outdoor datasets. The results show that our ensemble of active and passive outperforms the original RTAB-Map. Furthermore, the ensemble achieves a recall of 91.59{\%} and 98.65{\%} on the Lib6Indoor and Lib6Outdoor respectively, with a corresponding precision of 100{\%}.",
author = "Salameh, {Mohammed Omar} and Azizi Abdullah and Shahnorbanun Saran",
year = "2017",
month = "8",
day = "30",
doi = "10.1109/ICAR.2017.8023653",
language = "English",
pages = "482--486",
booktitle = "2017 18th International Conference on Advanced Robotics, ICAR 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Ensemble of Bayesian filter with active and passive nodes for loop closure detection

AU - Salameh, Mohammed Omar

AU - Abdullah, Azizi

AU - Saran, Shahnorbanun

PY - 2017/8/30

Y1 - 2017/8/30

N2 - In this paper, we describe a novel extension of the real-time appearance-based mapping (RTAB-Map), called the Ensemble of Real-Time Appearance-Based Mapping (ERTAB-Map). The original RTAB-Map calculates the probabilities of multiple beliefs for loop closure detection based on a single descriptor model. However, the ERTAB-Map can use an arbitrary number of descriptor models, in which a set of probability belief models are evaluated using an ensemble learning approach. The probability values are extracted from the active working memory and the passive long term memory of RTAB-Map. We have performed experiments on 388 images from the Lib6Indoor and 1063 images from Lib6Outdoor datasets. The results show that our ensemble of active and passive outperforms the original RTAB-Map. Furthermore, the ensemble achieves a recall of 91.59% and 98.65% on the Lib6Indoor and Lib6Outdoor respectively, with a corresponding precision of 100%.

AB - In this paper, we describe a novel extension of the real-time appearance-based mapping (RTAB-Map), called the Ensemble of Real-Time Appearance-Based Mapping (ERTAB-Map). The original RTAB-Map calculates the probabilities of multiple beliefs for loop closure detection based on a single descriptor model. However, the ERTAB-Map can use an arbitrary number of descriptor models, in which a set of probability belief models are evaluated using an ensemble learning approach. The probability values are extracted from the active working memory and the passive long term memory of RTAB-Map. We have performed experiments on 388 images from the Lib6Indoor and 1063 images from Lib6Outdoor datasets. The results show that our ensemble of active and passive outperforms the original RTAB-Map. Furthermore, the ensemble achieves a recall of 91.59% and 98.65% on the Lib6Indoor and Lib6Outdoor respectively, with a corresponding precision of 100%.

UR - http://www.scopus.com/inward/record.url?scp=85031662557&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85031662557&partnerID=8YFLogxK

U2 - 10.1109/ICAR.2017.8023653

DO - 10.1109/ICAR.2017.8023653

M3 - Conference contribution

SP - 482

EP - 486

BT - 2017 18th International Conference on Advanced Robotics, ICAR 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -