The first part is vision system which would detect the pad on the ground and compute the relative distance. This is based on opencv aruco marker algorithm. (/catkin_ws/src/opencvtest_py/src/test.py) The input of the system is the raw image of the camera and the output is the relative distance between drone and the launch pad in camera reference frame.

The second part is the PID control system(catkin_ws/control/src/control.py). When the vision system publishes the relative distance , the control system would subscribe it and use it as the input. A 3 dimension velocity loop PID control is built as we would control the x, y, z linear velocity to approach the mark.

cd ~/src/Firmware

source Tools/setup_gazebo.bash $(pwd) $(pwd)/build/px4_sitl_default export ROS_PACKAGE_PATH=$ROS_PACKAGE_PATH:$(pwd) export ROS_PACKAGE_PATH=$ROS_PACKAGE_PATH:$(pwd)/Tools/sitl_gazebo

roslaunch px4 drone.launch

rosrun offboard offboard_node

rosrun keyboard keyboard_ndoe

rosrun opencvtest_py test.py

rosrun control control.py