I attach the plot of a simulated flight (left) and of the mapping in lat/lon (left). In the simulation, the flight follows a path that only changes on the X-axis (straight line). All axii are in meters.
If X is longitude and Y is latitude, why is the behavior of the direct solution so erratic? (at least at high altitudes). I'm computing the azimuth in degrees clockwise from the Y+ axis.
I'm using the direct solution of this toolbox:
http://www.mathworks.com/matlabcentral/fileexchange/15285-geodetic-toolbox
The results are similar when using the following aviation formula:
http://williams.best.vwh.net/avform.htm#LL
Any insights?