This is a nice paper though I think for me quite worthless.
It does not go into enough detail around the aerodynamics of yawing the kite. It is clear that a kite can only yaw so much before performance degrades. Also, how much control input on rudder, aileron and flaps, and which associated drag and design constraints (how big flaps?)
If you want to optimize the trajectory before settling these details, you end up with very tight turns, because these are zero cost.
I think the better path forward is to gather real life data first, before trying to tackle this problem. Because I don’t think the analysis at the current state will give a useful answer.
A first step is going to this Betz limit and power available in the wind - #29 by tallakt
…but then also quite useless unless we check for stall, as this approach assumes linear lift coefficient with angle of attack.
Anyways, one could get closer to a really useful result.