Two-bounce LiDAR Digital Twin based data generation framework for object detection in the blind spot (Two-Bounce LiDAR와 디지털 트윈을 활용한 사각지대 객체 탐지용 데이터 생성 프레임워크)

Jae-Hun Hwang (황재훈); Seung Yeop Ha (하승엽); Jun-Seok Yun (윤준석); Min Su Kim (김민수); Sanga Lee (이상아); Jong Pil Yun (윤종필); Hong-In Won (원홍인)

doi:10.9708/jksci.2025.30.06.001

Two-bounce LiDAR Digital Twin based data generation framework for object detection in the blind spot

Journal of The Korea Society of Computer and Information
Abbr : JKSCI
2025, 30(6), pp.1~10
DOI : 10.9708/jksci.2025.30.06.001
Publisher : The Korean Society Of Computer And Information
Research Area : Engineering > Computer Science
Received : May 7, 2025
Accepted : May 29, 2025
Published : June 30, 2025

Jae-Hun Hwang ¹, Seung Yeop Ha ², Jun-Seok Yun ², Min Su Kim ², Sanga Lee ³, Jong Pil Yun ², Hong-In Won ²

¹한국생산기술연구원/한양대학교
²한국생산기술연구원
³한국생산기술연구원(인천)

Accredited

ABSTRACT

Obstacle detection in blind spots has emerged as a critical challenge due to the increasing adoption of Automated Guided Vehicle (AGV) and Autonomous Mobile Robot (AMR) systems in modern industrial environments. While conventional ultrasonic and single-bounce LiDAR sensors exhibit limitations in complex environments, Two-Bounce LiDAR offers higher detection precision but faces challenges such as high cost and limited diversity in training data. To address these challenge, this study proposes a framework that combines Two-Bounce LiDAR with Digital Twin simulation to generate synthetic data under various scenarios. Object detection models trained with this data achieved an average detection accuracy of 92.5% in environments including blind spots, with an average improvement of 14.67 percentage points in accuracy and 15.69 points in F1-score compared to models trained only on normal-environment data. These findings demonstrate that the proposed framework effectively overcomes data collection limitations and significantly enhances the safety and reliability of AGV and AMR operations.

KEYWORDS

Digital Twin, Data generation, Two-Bounce LiDAR, Blind spot, Object detection

Journal Copyright Policy

No CCL information provided

Citation status

* References for papers published after 2025 are currently being built.

[journal] Y. Bekishev / 2023 / FMEA model in risk analysis for the implementation of AGV/AMR robotic technologies into the internal supply system of enterprises / Risks 11(6) : 1~31

[confproc] M. Boehning / 2014 / Improving safety and efficiency of AGVs at warehouse black spots / Proc. IEEE 10th Int. Conf. Intell. Comput. Commun. Process. (ICCP) : 245~249

[journal] M. Wenning / 2022 / Testing the reliability of monocular obstacle detection methods in a simulated 3D factory environment / J. Intell. Manuf. 33(7) : 2157~2165

[journal] H.-C. Yoon / 2015 / Auto-parking Controller of Omnidirectional Mobile Robot Using Image Localization Sensor and Ultrasonic Sensors / J. Inst. Control, Robot. Syst. 21(6) : 571~576

[journal] Z. Qiu / 2022 / Review of ultrasonic ranging methods and their current challenges / Micromachines 13(520) : 1~33

[confproc] T. Klinghoffer / 2024 / PlatoNeRF: 3D reconstruction in Plato's cave via single-view two-bounce LiDAR / Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR) : 14565~14574

[journal] D. Lee / 2024 / LiDAR odometry survey: recent advancements and remaining challenges / Intell. Serv. Robot. 17(2) : 95~118

[journal] R. Geng / 2022 / Recent advances on non-line-of-sight imaging: Conventional physical models, deep learning, and new scenes / APSIPA Trans. Signal Inf. Process. 11(1) : 1~48

[journal] B. Wang / 2024 / Human digital twin in the context of Industry 5.0 / Rob. Comput.-Integr. Manuf. 85 : 102626~

[journal] S.-H. Shin / 2021 / Design of an Exploration Drone for Digital Twin based Building Control / J. Korea Soc. Comput. Inf. 26(5) : 9~16

[journal] S. Mihai / 2022 / Digital twins: A survey on enabling technologies, challenges, trends and future prospects / IEEE Commun. Surveys Tuts. 24(4) : 2255~2291

[confproc] A. Dosovitskiy / 2017 / CARLA: An open urban driving simulator / Proc. Conf. Robot Learn. (CoRL) / PMLR : 1~16

[book] S. Shah / 2018 / Airsim: High-fidelity visual and physical simulation for autonomous vehicles / Springer : 621~635

[confproc] D. A. Guerra-Zubiaga / 2019 / Digital twin in a manufacturing integrated system: Siemens TIA and PLM case study / ASME Int. Mech. Eng. Congr. Expo. 59384

[journal] N. V. Zorchenko / 2024 / Technologies used by General Electric to create digital twins for energy industry / Power Technol. Eng. 58(3) : 521~526

[journal] C. Henley / 2022 / Bounce-flash LiDAR / IEEE Trans. Comput. Imaging 8 : 411~424

[journal] S. Scholes / 2024 / Path tracing-inspired modeling of non-line-of-sight SPAD data / Sensors 24(20)

[confproc] M. Sandler / 2018 / MobileNetV2: Inverted residuals and linear bottlenecks / Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR) : 4510~4520

[other] F. Iandola / 2016 / SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size

[confproc] M. Tan / 2019 / EfficientNet: Rethinking model scaling for convolutional neural networks / Proc. Int. Conf. Mach. Learn. (ICML) : 6105~6114

KJCKorea
Journal Central

Journal of The Korea Society of Computer and Information 2025 KCI Impact Factor : 1.01

Two-bounce LiDAR Digital Twin based data generation framework for object detection in the blind spot

ABSTRACT

KEYWORDS

Journal Copyright Policy

Citation status

* References for papers published after 2025 are currently being built.

Journal of The Korea Society of Computer and Information 2025 KCI Impact Factor : 1.01

Two-bounce LiDAR Digital Twin based data generation framework for object detection in the blind spot

ABSTRACT

KEYWORDS

Journal Copyright Policy

Statistics

Tools

Issue List

Citation status

KCI Citation Counts (0)

REFERENCES (20) * References for papers published after 2025 are currently being built.

Search PDF

Citation

* References for papers published after 2025 are currently being built.