LUSE-MSCKF: A Landmark-Utilizing Enhancement of MSCKF for Robust and Efficient Monocular Visual-Inertial System

Hanyue Li; Zhiming Gao; Chunyu Li; Junhui Liu; Jiayuan Shan; Jianan Wang

doi:10.1007/978-981-92-4392-1_58

LUSE-MSCKF: A Landmark-Utilizing Enhancement of MSCKF for Robust and Efficient Monocular Visual-Inertial System

Hanyue Li
, Zhiming Gao
, Chunyu Li
, Junhui Liu
, Jiayuan Shan
, Jianan Wang^*

^*Corresponding author for this work

School of Aerospace Engineering

Beijing Institute of Technology

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

Abstract

When GPS signals are unreliable, robots must achieve autonomous localization using onboard sensors. Due to the limited computational resources of unmanned systems and the need to allocate these resources across multiple tasks, it is essential to enhance localization efficiency while maintaining accuracy. This paper proposes a landmark optimization method based on null space projection. A residual model is constructed from visual-inertial measurements, and equivalent residual equations for the state and landmark information are respectively constructed via null space projection. Extended Kalman Filter (EKF) updates are then performed separately on the state and landmarks. Experimental results on the public dataset (TUM-VI) demonstrate that the proposed method achieves superior localization accuracy and computational efficiency compared to the representative MSCKF-based (Multi-State Constraint Kalman Filter) open-source algorithm OpenVINS.

Original language	English
Title of host publication	Proceedings of 2025 9th Chinese Conference on Swarm Intelligence and Cooperative Control - Swarm Decision and Planning Technologies
Editors	Xiaoduo Li, Yichen Li, Zhi Feng
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	721-731
Number of pages	11
ISBN (Print)	9789819243914
DOIs	https://doi.org/10.1007/978-981-92-4392-1_58
Publication status	Published - 2027
Event	9th Chinese Conference on Swarm Intelligence and Cooperative Control, CCSICC 2025 - Shanghai, China Duration: 31 Oct 2025 → 3 Nov 2025

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	1602 LNEE
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	9th Chinese Conference on Swarm Intelligence and Cooperative Control, CCSICC 2025
Country/Territory	China
City	Shanghai
Period	31/10/25 → 3/11/25

Keywords

Extended Kalman Filter
MSCKF
Null space projection
Visual-inertial navigation system

Access to Document

10.1007/978-981-92-4392-1_58

Cite this

Li, H., Gao, Z., Li, C., Liu, J., Shan, J., & Wang, J. (2027). LUSE-MSCKF: A Landmark-Utilizing Enhancement of MSCKF for Robust and Efficient Monocular Visual-Inertial System. In X. Li, Y. Li, & Z. Feng (Eds.), Proceedings of 2025 9th Chinese Conference on Swarm Intelligence and Cooperative Control - Swarm Decision and Planning Technologies (pp. 721-731). (Lecture Notes in Electrical Engineering; Vol. 1602 LNEE). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-92-4392-1_58

Li, Hanyue ; Gao, Zhiming ; Li, Chunyu et al. / LUSE-MSCKF : A Landmark-Utilizing Enhancement of MSCKF for Robust and Efficient Monocular Visual-Inertial System. Proceedings of 2025 9th Chinese Conference on Swarm Intelligence and Cooperative Control - Swarm Decision and Planning Technologies. editor / Xiaoduo Li ; Yichen Li ; Zhi Feng. Springer Science and Business Media Deutschland GmbH, 2027. pp. 721-731 (Lecture Notes in Electrical Engineering).

@inproceedings{a1cf3cf695654eaf84c244d019ad0671,

title = "LUSE-MSCKF: A Landmark-Utilizing Enhancement of MSCKF for Robust and Efficient Monocular Visual-Inertial System",

abstract = "When GPS signals are unreliable, robots must achieve autonomous localization using onboard sensors. Due to the limited computational resources of unmanned systems and the need to allocate these resources across multiple tasks, it is essential to enhance localization efficiency while maintaining accuracy. This paper proposes a landmark optimization method based on null space projection. A residual model is constructed from visual-inertial measurements, and equivalent residual equations for the state and landmark information are respectively constructed via null space projection. Extended Kalman Filter (EKF) updates are then performed separately on the state and landmarks. Experimental results on the public dataset (TUM-VI) demonstrate that the proposed method achieves superior localization accuracy and computational efficiency compared to the representative MSCKF-based (Multi-State Constraint Kalman Filter) open-source algorithm OpenVINS.",

keywords = "Extended Kalman Filter, MSCKF, Null space projection, Visual-inertial navigation system",

author = "Hanyue Li and Zhiming Gao and Chunyu Li and Junhui Liu and Jiayuan Shan and Jianan Wang",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2027.; 9th Chinese Conference on Swarm Intelligence and Cooperative Control, CCSICC 2025 ; Conference date: 31-10-2025 Through 03-11-2025",

year = "2027",

doi = "10.1007/978-981-92-4392-1\_58",

language = "English",

isbn = "9789819243914",

series = "Lecture Notes in Electrical Engineering",

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

pages = "721--731",

editor = "Xiaoduo Li and Yichen Li and Zhi Feng",

booktitle = "Proceedings of 2025 9th Chinese Conference on Swarm Intelligence and Cooperative Control - Swarm Decision and Planning Technologies",

address = "Germany",

}

Li, H, Gao, Z, Li, C, Liu, J , Shan, J & Wang, J 2027, LUSE-MSCKF: A Landmark-Utilizing Enhancement of MSCKF for Robust and Efficient Monocular Visual-Inertial System. in X Li, Y Li & Z Feng (eds), Proceedings of 2025 9th Chinese Conference on Swarm Intelligence and Cooperative Control - Swarm Decision and Planning Technologies. Lecture Notes in Electrical Engineering, vol. 1602 LNEE, Springer Science and Business Media Deutschland GmbH, pp. 721-731, 9th Chinese Conference on Swarm Intelligence and Cooperative Control, CCSICC 2025, Shanghai, China, 31/10/25. https://doi.org/10.1007/978-981-92-4392-1_58

LUSE-MSCKF: A Landmark-Utilizing Enhancement of MSCKF for Robust and Efficient Monocular Visual-Inertial System. / Li, Hanyue; Gao, Zhiming; Li, Chunyu et al.
Proceedings of 2025 9th Chinese Conference on Swarm Intelligence and Cooperative Control - Swarm Decision and Planning Technologies. ed. / Xiaoduo Li; Yichen Li; Zhi Feng. Springer Science and Business Media Deutschland GmbH, 2027. p. 721-731 (Lecture Notes in Electrical Engineering; Vol. 1602 LNEE).

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

TY - GEN

T1 - LUSE-MSCKF

T2 - 9th Chinese Conference on Swarm Intelligence and Cooperative Control, CCSICC 2025

AU - Li, Hanyue

AU - Gao, Zhiming

AU - Li, Chunyu

AU - Liu, Junhui

AU - Shan, Jiayuan

AU - Wang, Jianan

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2027.

PY - 2027

Y1 - 2027

N2 - When GPS signals are unreliable, robots must achieve autonomous localization using onboard sensors. Due to the limited computational resources of unmanned systems and the need to allocate these resources across multiple tasks, it is essential to enhance localization efficiency while maintaining accuracy. This paper proposes a landmark optimization method based on null space projection. A residual model is constructed from visual-inertial measurements, and equivalent residual equations for the state and landmark information are respectively constructed via null space projection. Extended Kalman Filter (EKF) updates are then performed separately on the state and landmarks. Experimental results on the public dataset (TUM-VI) demonstrate that the proposed method achieves superior localization accuracy and computational efficiency compared to the representative MSCKF-based (Multi-State Constraint Kalman Filter) open-source algorithm OpenVINS.

AB - When GPS signals are unreliable, robots must achieve autonomous localization using onboard sensors. Due to the limited computational resources of unmanned systems and the need to allocate these resources across multiple tasks, it is essential to enhance localization efficiency while maintaining accuracy. This paper proposes a landmark optimization method based on null space projection. A residual model is constructed from visual-inertial measurements, and equivalent residual equations for the state and landmark information are respectively constructed via null space projection. Extended Kalman Filter (EKF) updates are then performed separately on the state and landmarks. Experimental results on the public dataset (TUM-VI) demonstrate that the proposed method achieves superior localization accuracy and computational efficiency compared to the representative MSCKF-based (Multi-State Constraint Kalman Filter) open-source algorithm OpenVINS.

KW - Extended Kalman Filter

KW - MSCKF

KW - Null space projection

KW - Visual-inertial navigation system

UR - https://www.scopus.com/pages/publications/105050801281

U2 - 10.1007/978-981-92-4392-1_58

DO - 10.1007/978-981-92-4392-1_58

M3 - Conference contribution

AN - SCOPUS:105050801281

SN - 9789819243914

T3 - Lecture Notes in Electrical Engineering

SP - 721

EP - 731

BT - Proceedings of 2025 9th Chinese Conference on Swarm Intelligence and Cooperative Control - Swarm Decision and Planning Technologies

A2 - Li, Xiaoduo

A2 - Li, Yichen

A2 - Feng, Zhi

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 31 October 2025 through 3 November 2025

ER -

Li H, Gao Z, Li C, Liu J , Shan J , Wang J. LUSE-MSCKF: A Landmark-Utilizing Enhancement of MSCKF for Robust and Efficient Monocular Visual-Inertial System. In Li X, Li Y, Feng Z, editors, Proceedings of 2025 9th Chinese Conference on Swarm Intelligence and Cooperative Control - Swarm Decision and Planning Technologies. Springer Science and Business Media Deutschland GmbH. 2027. p. 721-731. (Lecture Notes in Electrical Engineering). doi: 10.1007/978-981-92-4392-1_58

LUSE-MSCKF: A Landmark-Utilizing Enhancement of MSCKF for Robust and Efficient Monocular Visual-Inertial System

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this