Can't save solutions when using path constraint

[Hojin Song]

Hi Nick,

Thank you for the help! Just wondering, since I still wish to see the unconverged results, would there be any detour that I can get those failed results other than using unseal() (as it's not working with a path constraint)?

[Edward Syrett]

I don't have a solution to this, but I wanted to add that I saw a similar issue when adding a FrameDistanceConstraint. If I include that constraint, I cannot unseal a failed solution. I receive the following error message.

Java exception occurred:
java.lang.RuntimeException: casVector should be 1-dimensional, but has size 0 x 0.
Thrown at MocoCasOCProblem.h:108 in convertToSimTKVector().

at org.opensim.modeling.opensimMocoJNI.MocoStudy_solve(Native Method)

at org.opensim.modeling.MocoStudy.solve(MocoStudy.java:270)

Using identical code and commenting out the lines containing FrameDistanceConstraint, I can successfully unseal a non-converged solution.

[Nicholas Bianco]

Thanks Edward -- it seems like there is a bug related to problems that fail when using path constraints. Creating a GitHub issue now to capture this.

[Matthew Lee]

Hi Nick,

I also encountered a problem like what Hojin pointed.When I use C++ to solve a problem which including the path constraint,if the convergence tolerance is not satisfied while the the Maximum Number of Iterations is acheived,the program will be interrupted and the Exception will be thrown as follows:
Microsoft C++ exception:fmt::v6::format_error at memory location

I also find that if I use

Code: Select all

solver.set_output_interval(1)

,the solver does not even perform one step iterative calculation.

Could you give my some advice about how to resolve it so that the program will continue running and then I can unseal the failed solution?I can build the source code to fix this if needed.Thanks for your patience and time.

Best Wishes,
Matthew

[Nicholas Bianco]

Hi Matthew,

Thanks for reminding me about this issue. It is high on our development priority list, I will look into it soon.

Best,
Nick