I think geofences are typically thought of as static, although of course they don't have to be. They are usually known at mission-planning time, rather than discovered during the mission (not generally calculated based on Sensory data, although of course they could be).
Obstacles need to be detected and avoided during a mission. Detection is the hard part. If the obstacle is also moving, avoiding it might require trajectory calculations that are not required for simple geofence rules. It might also involve an internal model of expected behaviour of the obstacle (it's a moth, it will probably fly towards the light. It's a jumbo jet, it probably won't get out of my way).
What is the status of the old TCAS code? I never really looked at it, but think it might have done the trajectory calculations / avoidance part.
Chris Gough
On 5 Mar 2016, at 6:16 PM, Hector Garcia de Marina < address@hidden> wrote: Hi Andreas,
I am curious why they address geofencing the same as obstacle avoidance.
What (many) have in mind is that the former is to confine your mission's area whereas the latter is about avoiding collisions with nearby objects.
|