Dear Joris,
I was wondering how it might be possible to incorporate a height map, where the height
at a position on a grid is given by a function such as z = height(x, y) within a trajectory
optimization framework for a robot. Because this function only takes real values for x, and y, and not casadi MX or SX variables, how could I evaluate z in this case? (I am assuming x and y are casadi symbolic variables, since I won't know before hand where the robot will be, but I do know, using the height(x,y) function, what the z position would be at that position).
Sincerely,
Alex