• 设为首页
  • 点击收藏
  • 手机版
    手机扫一扫访问
    迪恩网络手机版
  • 关注官方公众号
    微信扫一扫关注
    迪恩网络公众号

qintonguav/Co-VINS: Collaborative Localization for Multiple Monocular Visual-Ine ...

原作者: [db:作者] 来自: 网络 收藏 邀请

开源软件名称(OpenSource Name):

qintonguav/Co-VINS

开源软件地址(OpenSource Url):

https://github.com/qintonguav/Co-VINS

开源编程语言(OpenSource Language):

C++ 95.1%

开源软件介绍(OpenSource Introduction):

Co-VINS

Collaborative Localization for Multiple Monocular Visual-Inertial Systems

Co-VINS a collaborative localization framework for multiple robots using monocular visual-inertial systems (VINS). Unlike traditional swarm applications which rely on external position equipment (GPS or Motion Capture System), our system achieves globally consistent localization based on internal sensors (onboard camera and IMU). Each robot is equipped with one camera and one IMU. It estimates own pose onboard and sends visual information to a centralized ground station. The ground station collects collaborative information from all robots, and maintains a globally consistent coordinate by pose graph optimization. Then the global localization is feedbacked to each robot for global control purpose. The ground station not only aligns all robots in a global coordinate, but also correct accumulated drifts for each robot. Co-VINS is an extension of VINS-Mono.

Authors: Tong Qin, William Wu, and Shaojie Shen from HUKST Aerial Robotics Group

Videos:

Swarm application

Related Papers

  • Technical report: Collaborative Localization for Multiple Monocular Vision-Based MAVs, Tong Qin, William Wu, Shaojie Shen Technical report

1. Prerequisites

1.1 Ubuntu and ROS Ubuntu 16.04. ROS Kinetic. ROS Installation additional ROS pacakge

    sudo apt-get install ros-YOUR_DISTRO-cv-bridge ros-YOUR_DISTRO-tf ros-YOUR_DISTRO-message-filters ros-YOUR_DISTRO-image-transport

1.2. Ceres Solver Follow Ceres Installation, remember to make install. (Our testing environment: Ubuntu 16.04, ROS Kinetic, OpenCV 3.3.1, Eigen 3.3.3)

2. Build Co-VINS on ROS

Clone the repository and catkin_make: (If VINS-Mono exists in your workspace, please remove it first)

    cd ~/catkin_ws/src
    git clone https://github.com/qintonguav/Co-VINS.git
    cd ../
    catkin_make
    source ~/catkin_ws/devel/setup.bash

3. Run datasets

Use your own computer as both multiple robots and ground station. Run several VIOs and pose graph optimization at the same time. (If you cannot run it in real time, it's ok. Because your computer serves as both multiple robots and ground station.)

3.1 Euroc dataset

3.1.1 Download EuRoC MAV Dataset. Although it contains stereo cameras, we only use one camera. We run multiple sequences simultaneously and merge them in one global frame.

3.1.2 Modify the dataset path in the launch file (vins_estimator/launch/euroc_multi_agent.launch)

<arg name="sequence_1" default = "YOUR_PATH/MH_01_easy/MH_01_easy.bag" />
<arg name="sequence_2" default = "YOUR_PATH/MH_02_easy/MH_02_easy.bag" />
<arg name="sequence_3" default = "YOUR_PATH/MH_03_medium/MH_03_medium.bag" />

3.1.3 Open terminal

roslaunch vins_estimator euroc_multi_agent.launch

euroc result

3.2 Our data (video)

Reproduction of part II in video.

3.2.1 Download Our Data.

3.2.2 Modify the dataset path in the launch file (vins_estimator/launch/A3_swarm.launch)

<arg name="sequence_1" default = "YOUR_PATH/1.bag" />
<arg name="sequence_2" default = "YOUR_PATH/2.bag" />
<arg name="sequence_3" default = "YOUR_PATH/3.bag" />
<arg name="sequence_4" default = "YOUR_PATH/4.bag" />

3.2.3 Open terminal

roslaunch vins_estimator A3_swarm.launch

(If it gose well, you can reproduce the part II in video. ) swarm result

4. Acknowledgements

We use ceres solver for non-linear optimization and DBoW2 for loop detection, and a generic camera model.

5. License

The source code is released under GPLv3 license.

We are still working on improving the code reliability. For any technical issues, please contact Tong QIN <tong.qinATconnect.ust.hk>.

For commercial inquiries, please contact Shaojie SHEN <eeshaojieATust.hk>




鲜花

握手

雷人

路过

鸡蛋
该文章已有0人参与评论

请发表评论

全部评论

专题导读
热门推荐
阅读排行榜

扫描微信二维码

查看手机版网站

随时了解更新最新资讯

139-2527-9053

在线客服(服务时间 9:00~18:00)

在线QQ客服
地址:深圳市南山区西丽大学城创智工业园
电邮:jeky_zhao#qq.com
移动电话:139-2527-9053

Powered by 互联科技 X3.4© 2001-2213 极客世界.|Sitemap