Authors: He Yijia, Gao Xiang, Cui huakun, Zhao Song
Introduction: This project is used for an online course about visual inertial odometry on shenlanxueyuan. She is based on the VINS-Mono framework but does not rely on ROS, Ceres, G2o. This code is very basic and aims to demonstrate Eigen-based back-end LM algorithms, sliding window algorithms, robust kernel functions, etc. This code supports for Ubuntu or Mac OS.
-
opencv
-
Eigen
-
Ceres. Uhhhhh, vins-mono use ceres to do sfm bundle adjustment. So we need ceres.
mkdir vins_course
cd vins_course
git clone https://github.com/HeYijia/VINS-Course
mkdir build
cd build
cmake ..
make -j4
cd build
../bin/testCurveFitting
cd build
../bin/run_euroc /home/dataset/EuRoC/MH-05/mav0/ ../config/
you can use this code to generate vio data.
https://github.com/HeYijia/vio_data_simulation
evo_ape euroc euroc_mh05_groundtruth.csv pose_output.txt -a -p
The source code is released under GPLv3 license.
We are still working on improving the code reliability. For any technical issues, please contact Yijia He [email protected] , Xiang Gao https://github.com/gaoxiang12 or Huakun Cuihttps://github.com/StevenCui.
For commercial inquiries, please contact Song Zhao [email protected]
We use VINS-Mono as our base line code. Thanks Dr. Qin Tong, Prof. Shen etc very much.