The codes works for AR.drone 2.0 or 1.0, we assume that you already have a new one. However, if you couldn't buy it temporarily, it's available that using a simulator(http://wiki.ros.org/tum_simulator) published by TUM Robotic Group to simulate the whole process.

OpenCV:https://opencv.org/ (reconmended version: less than 3.0.0)

Egien:http://eigen.tuxfamily.org/index.php?title=Main_Page

g2o: https://github.com/RainerKuemmerle/g2o

PCL:https://github.com/PointCloudLibrary/pcl (1.8)

ardrone_autonomy(edited verison:https://github.com/tum-vision/ardrone_autonomy or original version:http://wiki.ros.org/ardrone_autonomy)

you also have to install other dependencies used for each libraries. In addition, you should create a new folder named Thirdparty under SLAM and put g2o and DBoW2 into it. Due to the limitation of the amount of uploading files, I have to upload g2o and DBow2 separately.

ros-indigo + Ubuntu 14.04(best choice)

ros-kinetic+Ubuntu 16.04(probably good)

please build up a rosbuild(http://wiki.ros.org/rosbuild) workspace to store these packages, if you want to use catkin, please edit and create relevant files.

It is not uncommon that you fail to compile one of packages and build up new executable files. If you really find some bugs , you could send an email to editers and original author directly.

We use g2o for non-linear optimization, DBoW2 for loop detection, and ORBSLAM2 for feature extraction.

The source code is released under GPLv3 license.

We are still working on improving the code reliability. For any technical issues, please contact Omid [email protected], Mouaad [email protected], Tengfei Han[email protected].