When head tracking is activated, does the yaw resulting from turning your head do a simple rotation of the aircraft altering its course direction, or does it instead simulate a full 3-axis gimbal and decouple heading from course so you can "look around" while not affecting the direction the aircraft is traveling?
I suspect the former (bad!), even though the Mavic is way beyond capable of the latter with the sophisticated, redundant, sensitive sensor package it has.
For example, this is exactly what "headless mode" is on cheap toy drones. It only requires a six-axis accelerometer, which are a dime a dozen these days and the Mavic has two very good precision IMUs.
Backed up with a periodic correction from GPS data (regression over some number of samples), and the Mavic could easily calculate and maintain course independent of where the "front" of the quad is pointing. Actual course direction changes through left stick yaw input would just simulate the yaw in software, while controlling the quad to follow a curved course.
Both "modes" should be available.
I suspect the former (bad!), even though the Mavic is way beyond capable of the latter with the sophisticated, redundant, sensitive sensor package it has.
For example, this is exactly what "headless mode" is on cheap toy drones. It only requires a six-axis accelerometer, which are a dime a dozen these days and the Mavic has two very good precision IMUs.
Backed up with a periodic correction from GPS data (regression over some number of samples), and the Mavic could easily calculate and maintain course independent of where the "front" of the quad is pointing. Actual course direction changes through left stick yaw input would just simulate the yaw in software, while controlling the quad to follow a curved course.
Both "modes" should be available.
