Meijun Guo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Meijun Guo has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Weimin Zhang
School of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, China
4 shared publications - Chandan Sheikder
School of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, China
2 shared publications - Chengxiang Li
School of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, China
2 shared publications - Shicheng Fan
School of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, China
2 shared publications - Xiaohai He
School of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, China
2 shared publications - Fangxing Li
School of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, China
1 shared publication
Latest Publications
- Open Access
- 12 August 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (16) , 8028
LV-GS SLAM: A Decoupled LiDAR–Visual 3D Gaussian Splatting SLAM in a Large-Scale Environment
3D Gaussian Splatting (3DGS) has gained prominence in autonomous driving and robotics for its rendering efficiency and high-fidelity reconstruction capabilities. However, incremental 3DGS map construction at large scales remains challenging due to sensor sparsity and computational constraints. In this paper, we propose LV-GS SLAM, a novel system that integrates LiDAR and visual data for incremental, large-scale reconstruction with real-time tracking. This system provides fast and robust pose estimation while also enabling photorealistic rendering. We first employ a LiDAR odometry frontend that processes 30 Hz LiDAR inputs to provide robust initial poses. In our implementation, the complete LiDAR tracking pipeline runs at 15–19 Hz, while the mapping module performs incremental optimization on selected keyframes. To address the sparsity-induced surface discontinuity in conventional LiDAR-based reconstruction, we propose a novel depth propagation approach that initializes 3D Gaussian primitives using dense depth maps, achieving faster PSNR convergence with 9× fewer optimization iterations compared to direct LiDAR initialization. Furthermore, we develop a keyframe-based submap management framework that dynamically adjusts memory allocation based on both primitive density and inter-frame overlap ratio, effectively preventing GPU memory overflow. Our system has been validated on the KITTI dataset, achieving superior rendering quality compared with representative reproducible baselines. We further validate the robustness of the system on a quadruped robot platform, demonstrating satisfactory performance in both pose estimation and high-fidelity reconstruction.
- 12 August 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (16) , 8028
Open AccessLV-GS SLAM: A Decoupled LiDAR–Visual 3D Gaussian Splatting SLAM in a Large-Scale Environment
3D Gaussian Splatting (3DGS) has gained prominence in autonomous driving and robotics for its rendering efficiency and high-fidelity reconstruction capabilities. However, incremental 3DGS map construction at large scales remains challenging due to sensor sparsity and computational constraints. In this paper, we propose LV-GS SLAM, a novel system that integrates LiDAR and visual data for incremental, large-scale reconstruction with real-time tracking. This system provides fast and robust pose estimation while also enabling photorealistic rendering. We first employ a LiDAR odometry frontend that processes 30 Hz LiDAR inputs to provide robust initial poses. In our implementation, the complete LiDAR tracking pipeline runs at 15–19 Hz, while the mapping module performs incremental optimization on selected keyframes. To address the sparsity-induced surface discontinuity in conventional LiDAR-based reconstruction, we propose a novel depth propagation approach that initializes 3D Gaussian primitives using dense depth maps, achieving faster PSNR convergence with 9× fewer optimization iterations compared to direct LiDAR initialization. Furthermore, we develop a keyframe-based submap management framework that dynamically adjusts memory allocation based on both primitive density and inter-frame overlap ratio, effectively preventing GPU memory overflow. Our system has been validated on the KITTI dataset, achieving superior rendering quality compared with representative reproducible baselines. We further validate the robustness of the system on a quadruped robot platform, demonstrating satisfactory performance in both pose estimation and high-fidelity reconstruction.
- Open Access
HV-LIOM: Adaptive Hash-Voxel LiDAR–Inertial SLAM with Multi-Resolution Relocalization and Reinforcement Learning for Autonomous Exploration
This paper presents HV-LIOM (Adaptive Hash-Voxel LiDAR–Inertial Odometry and Mapping), a unified LiDAR–inertial SLAM and autonomous exploration framework for real-time 3D mapping in dynamic, GNSS-denied environments. We propose an adaptive hash-voxel mapping scheme that improves memory efficiency and real-time state estimation by subdividing voxels according to local geometric complexity and point density. To enhance robustness to poor initialization, we introduce a multi-resolution relocalization strategy that enables reliable localization against a prior map under large initial pose errors. A learning-based loop-closure module further detects revisited places and injects global constraints, while global pose-graph optimization maintains long-term map consistency. For autonomous exploration, we integrate a Soft Actor–Critic (SAC) policy that selects informative navigation targets online, improving exploration efficiency in unknown scenes. We evaluate HV-LIOM on public datasets (Hilti and NCLT) and a custom mobile robot platform. Results show that HV-LIOM improves absolute pose accuracy by up to 15.2% over FAST-LIO2 in indoor settings and by 7.6% in large-scale outdoor scenarios. The learned exploration policy achieves comparable or superior area coverage with reduced travel distance and exploration time relative to sampling-based and learning-based baselines.
Open AccessHV-LIOM: Adaptive Hash-Voxel LiDAR–Inertial SLAM with Multi-Resolution Relocalization and Reinforcement Learning for Autonomous Exploration
This paper presents HV-LIOM (Adaptive Hash-Voxel LiDAR–Inertial Odometry and Mapping), a unified LiDAR–inertial SLAM and autonomous exploration framework for real-time 3D mapping in dynamic, GNSS-denied environments. We propose an adaptive hash-voxel mapping scheme that improves memory efficiency and real-time state estimation by subdividing voxels according to local geometric complexity and point density. To enhance robustness to poor initialization, we introduce a multi-resolution relocalization strategy that enables reliable localization against a prior map under large initial pose errors. A learning-based loop-closure module further detects revisited places and injects global constraints, while global pose-graph optimization maintains long-term map consistency. For autonomous exploration, we integrate a Soft Actor–Critic (SAC) policy that selects informative navigation targets online, improving exploration efficiency in unknown scenes. We evaluate HV-LIOM on public datasets (Hilti and NCLT) and a custom mobile robot platform. Results show that HV-LIOM improves absolute pose accuracy by up to 15.2% over FAST-LIO2 in indoor settings and by 7.6% in large-scale outdoor scenarios. The learned exploration policy achieves comparable or superior area coverage with reduced travel distance and exploration time relative to sampling-based and learning-based baselines.
- Open Access
- 27 June 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (13) , 2214
VOX-LIO: An Effective and Robust LiDAR-Inertial Odometry System Based on Surfel Voxels
Accurate and robust pose estimation is critical for simultaneous localization and mapping (SLAM), and multi-sensor fusion has demonstrated efficacy with significant potential for robotic applications. This study presents VOX-LIO, an effective LiDAR-inertial odometry system. To improve both robustness and accuracy, we propose an adaptive hash voxel-based point cloud map management method that incorporates surfel features and planarity. This method enhances the efficiency of point-to-surfel association by leveraging long-term observed surfel. It facilitates the incremental refinement of surfel features within classified surfel voxels, thereby enabling precise and efficient map updates. Furthermore, we develop a weighted fusion approach that integrates LiDAR and IMU data measurements on the manifold, effectively compensating for motion distortion, particularly under high-speed LiDAR motion. We validate our system through experiments conducted on both public datasets and our mobile robot platforms. The results demonstrate that VOX-LIO outperforms the existing methods, effectively handling challenging environments while minimizing computational cost.
- 27 June 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (13) , 2214
Open AccessVOX-LIO: An Effective and Robust LiDAR-Inertial Odometry System Based on Surfel Voxels
Accurate and robust pose estimation is critical for simultaneous localization and mapping (SLAM), and multi-sensor fusion has demonstrated efficacy with significant potential for robotic applications. This study presents VOX-LIO, an effective LiDAR-inertial odometry system. To improve both robustness and accuracy, we propose an adaptive hash voxel-based point cloud map management method that incorporates surfel features and planarity. This method enhances the efficiency of point-to-surfel association by leveraging long-term observed surfel. It facilitates the incremental refinement of surfel features within classified surfel voxels, thereby enabling precise and efficient map updates. Furthermore, we develop a weighted fusion approach that integrates LiDAR and IMU data measurements on the manifold, effectively compensating for motion distortion, particularly under high-speed LiDAR motion. We validate our system through experiments conducted on both public datasets and our mobile robot platforms. The results demonstrate that VOX-LIO outperforms the existing methods, effectively handling challenging environments while minimizing computational cost.
- Open Access
- 26 June 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (13) , 7186
Vision-Based Detection, Localization, and Optimized Path Planning for Rebar Intersections in Automated Construction
The accurate detection and precise spatial localization of rebar intersection points are essential for advancing automation in construction tasks, such as robotic rebar tying. This paper presents a vision-based methodology that integrates RGB-D sensing, camera calibration, and coordinate transformation techniques to robustly detect and localize rebar crossing points. A structured detection framework efficiently extracts intersection coordinates from RGB-D imagery, subsequently mapping these points to a global reference frame using extrinsic camera calibration parameters. To achieve comprehensive site coverage and optimize operational efficiency, the path planning challenge is reformulated as a sequencing optimization problem of the identified intersections. We propose a greedy optimization algorithm that generates smooth, snake-like traversal paths in an efficient manner. Experimental validation confirms the effectiveness of our approach, demonstrating detection accuracy exceeding 99%, an average processing time below 125 ms per intersection point, and a maximum coordinate transformation error under 2 mm. The presented solution offers a lightweight, precise, and scalable framework, significantly facilitating the integration of vision-based methods into automated construction workflows.
- 26 June 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (13) , 7186
Open AccessVision-Based Detection, Localization, and Optimized Path Planning for Rebar Intersections in Automated Construction
The accurate detection and precise spatial localization of rebar intersection points are essential for advancing automation in construction tasks, such as robotic rebar tying. This paper presents a vision-based methodology that integrates RGB-D sensing, camera calibration, and coordinate transformation techniques to robustly detect and localize rebar crossing points. A structured detection framework efficiently extracts intersection coordinates from RGB-D imagery, subsequently mapping these points to a global reference frame using extrinsic camera calibration parameters. To achieve comprehensive site coverage and optimize operational efficiency, the path planning challenge is reformulated as a sequencing optimization problem of the identified intersections. We propose a greedy optimization algorithm that generates smooth, snake-like traversal paths in an efficient manner. Experimental validation confirms the effectiveness of our approach, demonstrating detection accuracy exceeding 99%, an average processing time below 125 ms per intersection point, and a maximum coordinate transformation error under 2 mm. The presented solution offers a lightweight, precise, and scalable framework, significantly facilitating the integration of vision-based methods into automated construction workflows.