Comparison of ROS-Based Monocular Visual SLAM Methods: DSO, LDSO, ORB-SLAM2 and DynaSLAM

Eldar Mingachev; Roman Lavrenov; Tatyana Tsoy; Fumitoshi Matsuno; Mikhail Svinin; Jackrit Suthakorn; Evgeni Magid

doi:10.1007/978-3-030-60337-3_22

Comparison of ROS-Based Monocular Visual SLAM Methods: DSO, LDSO, ORB-SLAM2 and DynaSLAM

Eldar Mingachev
, Roman Lavrenov
, Tatyana Tsoy
, Fumitoshi Matsuno
, Mikhail Svinin
, Jackrit Suthakorn
, Evgeni Magid

Department of Biomedical Engineering

Kazan Federal University
Kyoto University
Ritsumeikan University Biwako-Kusatsu Campus

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

34 Citations (Scopus)

Abstract

Stable and robust path planning of a ground mobile robot requires a combination of accuracy and low latency in its state estimation. Yet, state estimation algorithms should provide these under computational and power constraints of a robot embedded hardware. The presented study offers a comparative analysis of four cutting edge publicly available within robot operating system (ROS) monocular simultaneous localization and mapping methods: DSO, LDSO, ORB-SLAM2, and DynaSLAM. The analysis considers pose estimation accuracy (alignment, absolute trajectory, and relative pose root mean square error) and trajectory precision of the four methods at TUM-Mono and EuRoC datasets.

Original language	English
Title of host publication	Interactive Collaborative Robotics - 5th International Conference, ICR 2020, Proceedings
Editors	Andrey Ronzhin, Gerhard Rigoll, Roman Meshcheryakov
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	222-233
Number of pages	12
ISBN (Print)	9783030603366
DOIs	https://doi.org/10.1007/978-3-030-60337-3_22
Publication status	Published - 2020
Event	5th International Conference on Interactive Collaborative Robotics, ICR 2020 - St. Petersburg, Russian Federation Duration: 7 Oct 2020 → 9 Oct 2020

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12336 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	5th International Conference on Interactive Collaborative Robotics, ICR 2020
Country/Territory	Russian Federation
City	St. Petersburg
Period	7/10/20 → 9/10/20

Keywords

Benchmark testing
Monocular SLAM
Path planning
Robot sensing systems
Simultaneous localization and mapping
State estimation
Visual SLAM
Visual odometry

Access to Document

10.1007/978-3-030-60337-3_22

Cite this

Mingachev, E., Lavrenov, R., Tsoy, T., Matsuno, F., Svinin, M., Suthakorn, J., & Magid, E. (2020). Comparison of ROS-Based Monocular Visual SLAM Methods: DSO, LDSO, ORB-SLAM2 and DynaSLAM. In A. Ronzhin, G. Rigoll, & R. Meshcheryakov (Eds.), Interactive Collaborative Robotics - 5th International Conference, ICR 2020, Proceedings (pp. 222-233). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12336 LNAI). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-60337-3_22

Mingachev, Eldar ; Lavrenov, Roman ; Tsoy, Tatyana et al. / Comparison of ROS-Based Monocular Visual SLAM Methods : DSO, LDSO, ORB-SLAM2 and DynaSLAM. Interactive Collaborative Robotics - 5th International Conference, ICR 2020, Proceedings. editor / Andrey Ronzhin ; Gerhard Rigoll ; Roman Meshcheryakov. Springer Science and Business Media Deutschland GmbH, 2020. pp. 222-233 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{5161cc9401034f8e8bfecabb020f8e61,

title = "Comparison of ROS-Based Monocular Visual SLAM Methods: DSO, LDSO, ORB-SLAM2 and DynaSLAM",

abstract = "Stable and robust path planning of a ground mobile robot requires a combination of accuracy and low latency in its state estimation. Yet, state estimation algorithms should provide these under computational and power constraints of a robot embedded hardware. The presented study offers a comparative analysis of four cutting edge publicly available within robot operating system (ROS) monocular simultaneous localization and mapping methods: DSO, LDSO, ORB-SLAM2, and DynaSLAM. The analysis considers pose estimation accuracy (alignment, absolute trajectory, and relative pose root mean square error) and trajectory precision of the four methods at TUM-Mono and EuRoC datasets.",

keywords = "Benchmark testing, Monocular SLAM, Path planning, Robot sensing systems, Simultaneous localization and mapping, State estimation, Visual SLAM, Visual odometry",

author = "Eldar Mingachev and Roman Lavrenov and Tatyana Tsoy and Fumitoshi Matsuno and Mikhail Svinin and Jackrit Suthakorn and Evgeni Magid",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG.; 5th International Conference on Interactive Collaborative Robotics, ICR 2020 ; Conference date: 07-10-2020 Through 09-10-2020",

year = "2020",

doi = "10.1007/978-3-030-60337-3\_22",

language = "English",

isbn = "9783030603366",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "222--233",

editor = "Andrey Ronzhin and Gerhard Rigoll and Roman Meshcheryakov",

booktitle = "Interactive Collaborative Robotics - 5th International Conference, ICR 2020, Proceedings",

}

Mingachev, E, Lavrenov, R, Tsoy, T, Matsuno, F, Svinin, M, Suthakorn, J & Magid, E 2020, Comparison of ROS-Based Monocular Visual SLAM Methods: DSO, LDSO, ORB-SLAM2 and DynaSLAM. in A Ronzhin, G Rigoll & R Meshcheryakov (eds), Interactive Collaborative Robotics - 5th International Conference, ICR 2020, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12336 LNAI, Springer Science and Business Media Deutschland GmbH, pp. 222-233, 5th International Conference on Interactive Collaborative Robotics, ICR 2020, St. Petersburg, Russian Federation, 7/10/20. https://doi.org/10.1007/978-3-030-60337-3_22

Comparison of ROS-Based Monocular Visual SLAM Methods: DSO, LDSO, ORB-SLAM2 and DynaSLAM. / Mingachev, Eldar; Lavrenov, Roman; Tsoy, Tatyana et al.
Interactive Collaborative Robotics - 5th International Conference, ICR 2020, Proceedings. ed. / Andrey Ronzhin; Gerhard Rigoll; Roman Meshcheryakov. Springer Science and Business Media Deutschland GmbH, 2020. p. 222-233 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12336 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Comparison of ROS-Based Monocular Visual SLAM Methods

T2 - 5th International Conference on Interactive Collaborative Robotics, ICR 2020

AU - Mingachev, Eldar

AU - Lavrenov, Roman

AU - Tsoy, Tatyana

AU - Matsuno, Fumitoshi

AU - Svinin, Mikhail

AU - Suthakorn, Jackrit

AU - Magid, Evgeni

PY - 2020

Y1 - 2020

N2 - Stable and robust path planning of a ground mobile robot requires a combination of accuracy and low latency in its state estimation. Yet, state estimation algorithms should provide these under computational and power constraints of a robot embedded hardware. The presented study offers a comparative analysis of four cutting edge publicly available within robot operating system (ROS) monocular simultaneous localization and mapping methods: DSO, LDSO, ORB-SLAM2, and DynaSLAM. The analysis considers pose estimation accuracy (alignment, absolute trajectory, and relative pose root mean square error) and trajectory precision of the four methods at TUM-Mono and EuRoC datasets.

AB - Stable and robust path planning of a ground mobile robot requires a combination of accuracy and low latency in its state estimation. Yet, state estimation algorithms should provide these under computational and power constraints of a robot embedded hardware. The presented study offers a comparative analysis of four cutting edge publicly available within robot operating system (ROS) monocular simultaneous localization and mapping methods: DSO, LDSO, ORB-SLAM2, and DynaSLAM. The analysis considers pose estimation accuracy (alignment, absolute trajectory, and relative pose root mean square error) and trajectory precision of the four methods at TUM-Mono and EuRoC datasets.

KW - Benchmark testing

KW - Monocular SLAM

KW - Path planning

KW - Robot sensing systems

KW - Simultaneous localization and mapping

KW - State estimation

KW - Visual SLAM

KW - Visual odometry

UR - https://www.scopus.com/pages/publications/85092904826

U2 - 10.1007/978-3-030-60337-3_22

DO - 10.1007/978-3-030-60337-3_22

M3 - Conference contribution

AN - SCOPUS:85092904826

SN - 9783030603366

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 222

EP - 233

BT - Interactive Collaborative Robotics - 5th International Conference, ICR 2020, Proceedings

A2 - Ronzhin, Andrey

A2 - Rigoll, Gerhard

A2 - Meshcheryakov, Roman

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 7 October 2020 through 9 October 2020

ER -

Mingachev E, Lavrenov R, Tsoy T, Matsuno F, Svinin M, Suthakorn J et al. Comparison of ROS-Based Monocular Visual SLAM Methods: DSO, LDSO, ORB-SLAM2 and DynaSLAM. In Ronzhin A, Rigoll G, Meshcheryakov R, editors, Interactive Collaborative Robotics - 5th International Conference, ICR 2020, Proceedings. Springer Science and Business Media Deutschland GmbH. 2020. p. 222-233. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-60337-3_22

Comparison of ROS-Based Monocular Visual SLAM Methods: DSO, LDSO, ORB-SLAM2 and DynaSLAM

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this