百科.dev
全部条目在线工具AI 编程趋势榜开源项目技术资讯提交条目
登录
< 返回工具列表
S

sqrtVINS

> 编程语言
开源

基于强大且超快速的平方根滤波器的 3D 运动跟踪

383 stars0 点赞1 次浏览
访问官网GitHub

工具介绍

基于强大且超快速的平方根滤波器的 3D 运动跟踪

${\sqrt{\rm VINS}}$

Welcome to the ${\sqrt{\rm VINS}}$ project! This repository provides the first open-source implementation of a robust and efficient square-root filter (SRF)-based visual-inertial navigation system (VINS), see arXiv. Extended from OpenVINS, unlike conventional EKF- or MSCKF-based systems, ${\sqrt{\rm VINS}}$ leverages a novel LLT-based SRF update to preserve the triangular structure of the covariance matrix, ensuring numerical stability, guaranteed positive semi-definiteness, and efficient memory usage. This makes it especially well-suited for embedded robotic platforms and systems constrained by limited precision.

Key highlights of ${\sqrt{\rm VINS}}$ include:

⚡ Ultra-fast and stable filtering – runs efficiently wih 32-bit single-precision on embedded hardware (e.g., Jetson Nano in 5W mode with single thread (<1 GHz CPU)).

Cholesky (LLT)-based SRF update – fully exploits system structure to achieve significant speedups over canonical SRFs.

Dynamic initialization – recovers minimal states without triangulating 3D features, enabling reliable startup even in extreme conditions (as small as a 100 ms time window).

Dependences

  • ROS
  • Eigen
  • OpenCV

Getting Started

Run in Native System with EurocMav Dataset (Ubuntu 20.04 + ROS1 as Example)

bash
# Step 1: Create the workspace
mkdir -p sqrt_vins_ws/src && cd sqrt_vins_ws/src
git clone https://github.com/rpng/sqrtVINS.git

# Step 2: Build 
catkin build

# Step 3: Download eurocmav rosbag(https://projects.asl.ethz.ch/datasets/doku.php?id=kmavvisualinertialdatasets) in $HOME/datasets/euroc_mav or change the path in /ov_srvins/launch/serial.launch accordingly

# Step 4: Run the launch file
source devel/setup.bash
roslaunch ov_srvins serial.launch

Docker Support

As the code base is extended from OpenVINS, the dockcer support can refer to OpenVINS's doc.

Generate Results with Scripts

Please refer to src/srf_init/ov_srvins/scripts/run_abtest.sh

Credit / Licensing

This code was written by the Robot Perception and Navigation Group (RPNG) at the University of Delaware. If you have any issues with the code please open an issue on our github page with relevant implementation details and references. For researchers that have leveraged or compared to this work, please cite the following:

…

The codebase is licensed under the GNU Lesser General Public License v3 (LGPL-3.0). You must preserve the copyright and license notices in your derivative work. If you modify the library itself, you must make those modifications available under the same license. You may link this library into your own programs (including proprietary software), provided that users are able to replace or relink with a modified version of the library. (See this; this is not legal advice.)

Issues· 0 开放

查看全部 Issues在 GitHub 打开

暂无开放 Issues,或尚未同步最近议题。

> 标签

C++

暂无评论,来聊聊你的看法吧

> 工具信息

发布日期2026年8月1日
最后更新2026年9月17日
分类编程语言
定价开源

> 相关工具

T
TypeScript
JavaScript 的超集,为前端与全栈提供静态类型
P
Python
通用编程语言,广泛用于 Web、数据与 AI
G
Go
Google 推出的简洁高效系统语言