College of Artificial Intelligence, Institute of Robotics and Automatic Information System, Tianjin Key Laboratory of Intelligent Robotics, Nankai University, Tianjin, China
Auto-generated profile, based on publicly available information
Yunze Tong
College of Artificial Intelligence, Institute of Robotics and Automatic Information System, Tianjin Key Laboratory of Intelligent Robotics, Nankai University, Tianjin, China
Robust and reliable calibration forms the foundation of efficient multi-sensor fusion. Most existing calibration methods are offline and rely on artificial targets, which is time consuming and unfriendly to non-expert users. To improve efficiency, robustness and user-friendliness, this paper proposes a novel target-free LiDAR-IMU-camera online extrinsic calibration framework. An IMU-centric motion excitation assessment method is newly designed to assist non-expert users in data collection, serving as the termination criterion for extrinsic rotation initialization. We calibrate the extrinsic parameters of the LiDAR and camera relative to the IMU in real-time and in parallel without any priori knowledge: (i) For camera-IMU calibration, a new convergence criterion based on the coefficient of variation is proposed to judge if the calibration has been accomplished. (ii) For LiDAR-IMU calibration, a data selection strategy is designed to concurrently record a segment of original data from both the LiDAR and IMU during initialization. Subsequently, the selected data is utilized to refine the extrinsic parameters under a continuous-time batch optimization framework. Moreover, a scene recognition algorithm is proposed to autonomously select parameters according to indoor or outdoor features. Finally, we leverage the IMU as a bridge for computing the LiDAR-camera extrinsic parameters. Extensive real experiments on our self-recorded and public datasets show the high accuracy and robustness of the proposed method, with an average error of less than 0.01m in translation and $0.5^{\circ }$ in rotation. (Supplementary video link: https://youtu.be/Iph0kd_LIlk )
10 July 2024 in IEEE Robotics and Automation Letters
In this letter, we propose TC $^{2}$ LI-SLAM, a novel and tightly-coupled camera-LiDAR-inertial SLAM system with a separated front end and a unified back end. The front end accepts non-identical frequency inputs and performs real-time pose estimation while incorporating different inputs into keyframes for the back end, benefiting from our novel keyframe-oriented data synchronization method. To improve the accuracy of the back-end optimization, we innovatively leverage multi-sensor consecutive co-vision constraints in a keyframe-based local bundle adjustment problem. To bridge the gap between different residual terms from different sensors in the optimization problem, we propose a cross-modal residual standardization method that can be easily applied. Furthermore, the proposed system can degrade into TC $^{2}$ L-SLAM, a tightly-coupled camera-LiDAR SLAM system without IMU, to cope with situations where inertial measurements are not available or reliable. Finally, both TC $^{2}$ LI-SLAM and TC $^{2}$ L-SLAM are tested on public and self-recorded datasets. The results show their superior performance in terms of accuracy compared with the state-of-the-art SLAM systems, with at most 33% accuracy improvement over ORB-SLAM3 on the KITTI dataset.
15 January 2026 in IEEE Robotics and Automation Letters
This letter proposes R-VoxelMap, a novel voxel mapping method that constructs accurate voxel maps using a geometry-driven recursive plane fitting strategy to enhance the localization accuracy of online LiDAR odometry. VoxelMap and its variants typically fit and check planes using all points in a voxel, which may lead to plane parameter deviation caused by outliers, over segmentation of large planes, and incorrect merging across different physical planes. To address these issues, R-VoxelMap utilizes a geometry-driven recursive construction strategy based on an outlier detect-and-reuse pipeline. Specifically, for each voxel, accurate planes are first fitted while separating outliers using random sample consensus (RANSAC). The remaining outliers are then propagated to deeper octree levels for recursive processing, ensuring a detailed representation of the environment. In addition, a point distribution-based validity check algorithm is devised to prevent erroneous plane merging. Extensive experiments on diverse open-source LiDAR(-inertial) simultaneous localization and mapping (SLAM) datasets validate that our method achieves higher accuracy than other state-of-the-art approaches, with comparable efficiency and memory usage.
1 September 2025 in IEEE Transactions on Aerospace and Electronic Systems
The fusion of LiDAR and camera data holds vast application prospects, with calibration as a crucial prerequisite. In this paper, we propose SPTL-LCC, a LiDAR -type agnostic and target-free LiDAR-camera extrinsic calibration framework. The core of SPTL-LCC lies in extracting accurate 3D-2D correspondences in 2D image space using a densified point cloud and an image. To ensure the acquisition of a densified point cloud, a point cloud accumulation evaluation metric is proposed. Building upon this, a multi-stage LiDAR image processing and optimization method based on a virtual camera is proposed. Then, multiple image matching networks are employed to estimate 2D-2D correspondences between the LiDAR images and camera images, resulting in 3D-2D correspondences. A progressive reprojection error minimization algorithm is subsequently proposed to optimize the extrinsic parameters based on these correspondences. SPTL-LCC improves LiDAR image quality in a few iterations, enhancing correspondence estimation accuracy, thus achieving coarse-to-fine calibration. Extensive experiments on four self-recorded datasets show SPTL-LCC outperforms state-of-the-art methods in terms of robustness, accuracy, and practicality, with an average error of less than 0.03 m in translation and 0.4$^{\circ }$ in rotation. Moreover, experiments on the KITTI dataset show SPTL-LCC even achieves comparable performance to end-to-end methods trained on this dataset, which exhibit poor generalization. Our code and datasets are open-source: https://github.com/NKU-MobFly-Robotics/SPTG-LCC.
5 December 2025 in IEEE Robotics and Automation Letters
Recent advancements in 3D Gaussian Splatting (3DGS) have significantly improved the mapping quality and computational efficiency of visual Simultaneous Localization and Mapping (SLAM). We propose DOGL-SLAM, a novel framework that integrates 3DGS into its core pipeline, enabling accurate camera pose tracking, object-level interaction, and high-fidelity scene reconstruction in dynamic environments. Firstly, a joint graph optimization module incorporates both dense Gaussian and sparse landmark constraints, enabling precise alignment between camera tracking and mapping. Secondly, a consistent object-level semantic fusion module embeds category labels into 3D Gaussians, grouping them via semantic consistency and optimizing distributions through cross-view losses to support scene manipulation. Finally, a hierarchical dynamic filtering pipeline is introduced, consisting of segmentation, dynamic feature exclusion, gradient masking, and visibility-aware Gaussian pruning across three parallel threads. Our system is evaluated across multiple datasets, showing a significant improvement in high-quality view synthesis for dynamic scenes. Additionally, the generated object-level semantic maps facilitate advanced downstream tasks, highlighting the potential of a robust SLAM framework. The source code and supplementary video will be released.