张宁 Robust High Accuracy Visual-Inertial-Laser SLAM System
强大的高精度视觉惯性激光SLAM系统
链接：https://pan.baidu.com/s/1hquneyozISab6tSCaKijCg
Zengyuan Wang, Jianhua Zhang, Shengyong Chen, Conger Yuan, Jingqian Zhang and Jianwei Zhang
提取码：v758
Abstract— In recent years, many excellent works on visual-inertial SLAM and laser-based SLAM have been proposed. Al-though inertial measurement unit (IMU) significantly improvethe motion estimate performance by reducing the impact ofillumination variation or texture-less region on visual tracking,tracking failures occur when in such an environment for along time. Similarly, when in structure-less environments, lasermodule will fail since lack of sufficient geometric features.Besides, motion estimation by moving lidar has the problem ofdistortion since range measurements are received continuously.To solve these problems, we propose a robust and high-accuracyvisual-inertial-laser SLAM system. The system starts with avisual-inertial tightly-coupled method for motion estimation,followed by scan matching to further optimize the estimationand register point cloud on the map. Furthermore, we enablemodules to be adjusted automatically and flexibly. That is,when one of these modules fails, the remaining modules will undertake the motion-tracking task. For further improving the accuracy, loop closure and proximity detection are implemented to eliminate drift accumulation. When loop or proximity isdetected, we perform six degree-of-freedom (6-DOF) pose graph optimization to achieve the global consistency. The performance of our system is verified on public dataset, and the experimental results show that the proposed method achieves superior accuracy against other state-of-the-art algorithms.
近年来，已经提出了许多关于视觉惯性SLAM和基于激光的SLAM的出色著作。尽管惯性测量单元（IMU）通过减少照明变化或无纹理区域对视觉跟踪的影响来显着提高运动估计性能，但在这种环境中长时间运行会导致跟踪失败。同样，在无结构的环境中，由于缺少足够的几何特征，激光模块也会发生故障。此外，由于连续接收距离测量值，因此通过移动激光雷达进行运动估计存在失真的问题。为了解决这些问题，我们提出了一个强大的，高精度的视觉惯性激光SLAM系统。 该系统从视觉-惯性紧密耦合的运动估计方法开始，然后进行扫描匹配以进一步优化估计和在地图上注册点云。此外，我们使模块能够自动灵活地进行调整。 也就是说，当这些模块之一发生故障时，其余模块将承担运动跟踪任务。为了进一步提高精度，实施了闭环和接近检测以消除漂移累积。 当检测到回路或接近度时，我们将执行六个自由度（6-DOF）姿态图优化，以实现全局一致性。 我们的系统性能在公共数据集上得到了验证，实验结果表明，与其他最新算法相比，该方法具有更高的精度。