I've changed the optimisation so, in terms of kinematics, it only tries to match the input data (which could be measured or desired), not the interpolated data (at each node).
For example, if we only specify the start and end of the movement, there will be no assumptions about the trajectory to get from start to end.
I've set the start and end velocity to be zero.
I've also included the changes in tolerances here, so I've deleted @RostiReadioff pull request (which contained output files we don't want under version control).
I've included two scripts, get_muscles_names.m and include_subset_muscles.m to make it easy to choose which muscles to include in the optimisation.
I've changed the optimisation so, in terms of kinematics, it only tries to match the input data (which could be measured or desired), not the interpolated data (at each node). For example, if we only specify the start and end of the movement, there will be no assumptions about the trajectory to get from start to end.
I've set the start and end velocity to be zero.
I've also included the changes in tolerances here, so I've deleted @RostiReadioff pull request (which contained output files we don't want under version control).
I've included two scripts, get_muscles_names.m and include_subset_muscles.m to make it easy to choose which muscles to include in the optimisation.