融合距离梯度的单目视觉-惯性-UWB紧耦合导航定位方法
CSTR:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

TP24

基金项目:

国家重点研发计划项目(2023YFB4704402);辽宁省科技厅项目(2023020427-JH1/104);国家自然科学基金项目(U22B2041);中科院青促会优秀会员项目(Y2022065).


Distance gradient integrated monocular visual-inertial-UWB tightly coupled localization approach
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    视觉惯性里程计(VIO)是卫星拒止环境下的高效导航定位方法, 但是由于传感器噪声和累计误差的存在, 不可避免地会产生漂移而影响导航定位精度. 针对上述问题, 提出一种融合距离梯度的单目视觉-惯性-UWB紧耦合导航定位方法. 通过构建超宽带(UWB)测距传感器的测量距离及其梯度信息的残差模型, 利用因子图优化方法实现VIO与UWB的信息融合, 特别是距离梯度信息的引入, 能够实现速度信息的修正, 从而以多传感器紧耦合融合的方式进一步抑制导航系统的漂移并改善导航定位精度. 将所提出方法在EuRoc公开数据集以及实际的无人机平台上进行多种飞行模式的实验验证, 实验结果表明: 所提出方法具有良好的可行性和导航定位性能, 与使用相同测量信息的同类方法比较, 导航精度能够提升13.7%.

    Abstract:

    Visual-inertial odometry(VIO) is an efficient localization method in GPS-denied environments. However, due to sensor noise and accumulated errors, drift inevitably occurs, affecting the accuracy of localization. To address this issue, this paper proposes a monocular visual-inertial-UWB tightly coupled localization method that integrates distance gradient information. By constructing a residual model of the distance and its gradient information from the ultra-wideband(UWB) ranging sensor, the information from VIO and UWB is fused using factor graph optimization. The introduction of distance gradient information helps to correct velocity, thereby further suppressing drift in the navigation system and improving localization accuracy through tight multi-sensor fusion. The proposed method is experimentally validated on the EuRoc public dataset and an actual UAV platform, covering various flight conditions. The results show that the proposed method is highly feasible and demonstrates superior navigation and localization performance, achieving a 13.7% improvement in accuracy compared to similar methods using the same measurement information.

    参考文献
    相似文献
    引证文献
引用本文

蒋浩然,谷丰,滕天启,等.融合距离梯度的单目视觉-惯性-UWB紧耦合导航定位方法[J].控制与决策,2025,40(8):2566-2578

复制
相关视频

分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2024-12-13
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2025-07-11
  • 出版日期: 2025-08-20
文章二维码