We shouldn't have any scaling problems if we only want to optimize trajectories with respect to the 10 payload parameters. Adding a Snopt version for this would enable rapid excitation trajectory design. This is important when we want to slightly adjust a given trajectory on the fly to achieve a bit more excitation.
Implementation wise, it should be possible to reuse lots of the existing Snopt experiment design code.
We shouldn't have any scaling problems if we only want to optimize trajectories with respect to the 10 payload parameters. Adding a Snopt version for this would enable rapid excitation trajectory design. This is important when we want to slightly adjust a given trajectory on the fly to achieve a bit more excitation.
Implementation wise, it should be possible to reuse lots of the existing Snopt experiment design code.