arXiv 2404.091692024-04-14cs.RO

通过地面-卫星图像配准提升SLAM位姿精度

Increasing SLAM Pose Accuracy by Ground-to-Satellite Image Registration

Robotics Institute, University of Technology Sydney（悉尼科技大学机器人研究所）；ShanghaiTech University（上海科技大学）；College of Engineering and Computer Science, Australian National University（澳大利亚国立大学工程与计算机科学学院）；Data61, CSIRO（联邦科学与工业研究组织Data61）；Ford Motor Company（福特汽车公司）

Yanhao Zhang, Yujiao Shi, Shan Wang, Ankit Vora, Akhil Perincherry, Yongbo Chen, Hongdong Li

备注：: 7 pages, 6 figures, to be published in 2024 International Conference on Robotics and Automation (ICRA)

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对视觉SLAM长期漂移问题，提出融合vSLAM与深度学习地面-卫星图像配准的框架，经粗细两级筛选有效预测、求解带尺度位姿图及迭代轨迹融合提升定位精度，在两个自动驾驶数据集上验证了准确性与鲁棒性。

摘要：

自动驾驶的基于视觉的定位技术一直是研究者关注的热点。当预构建的3D地图不可用时，通常采用视觉同步定位与地图构建（SLAM）技术。由于误差累积，视觉SLAM（vSLAM）通常存在长期漂移问题。本文提出一种框架，通过将vSLAM与基于深度学习的地面-卫星（G2S）图像配准方法融合来提升定位精度。该框架设计了从粗（空间相关性边界检查）到细（视觉里程计一致性检查）的方法来筛选有效的G2S预测结果，随后通过求解带尺度的位姿图问题将筛选后的预测结果与SLAM测量值融合。为进一步提升定位精度，本文还提出了一种迭代轨迹融合流程。该框架在两个知名自动驾驶数据集上进行了评估，结果证明了其在车辆定位方面的准确性和鲁棒性。

英文摘要：

Vision-based localization for autonomous driving has been of great interest among researchers. When a pre-built 3D map is not available, the techniques of visual simultaneous localization and mapping (SLAM) are typically adopted. Due to error accumulation, visual SLAM (vSLAM) usually suffers from long-term drift. This paper proposes a framework to increase the localization accuracy by fusing the vSLAM with a deep-learning-based ground-to-satellite (G2S) image registration method. In this framework, a coarse (spatial correlation bound check) to fine (visual odometry consistency check) method is designed to select the valid G2S prediction. The selected prediction is then fused with the SLAM measurement by solving a scaled pose graph problem. To further increase the localization accuracy, we provide an iterative trajectory fusion pipeline. The proposed framework is evaluated on two well-known autonomous driving datasets, and the results demonstrate the accuracy and robustness in terms of vehicle localization.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。