Position hold controller
Hi,
I am in the process of writing a position hold controller using visual odometry/optical flow but I am at a loss as to where to start. I currently have rpg_svo installed and have been able to calibrate my monocular camera to be used by the SVO package. I'm able to run SVO as well and can see the features being detected. I also have a Pixhawk flight controller that I will be using. So far, I've been looking at what kinds of outputs I can get from the SVO package and it looks like one of them would be geometry_msgs/Pose, which contains a point and an orientation. I am still a newbie in the drone world, but I think that I would need some way to translate this information into roll, pitch, and yaw for my flight controller. Is this the correct way of thinking about approaching this problem? Additionally, has anyone tried to use SVO for position hold and can provide me with some tutorials/sample code?
Thank you!
