Thanks a lot for developing such a wonderful toolbox. I am very new to OCP and trying to setup one in python interface but face some issues regarding following points.
- My model is defined by nonlinear equation f(x,u,p) where parameter p needs an update at each stage. Currently I defined them as P = MX(‘p’, np, 1) put them in model.p following the guidelines of the example of race cars. Further I also set nlp_dims.np = model.p.size() in acados_settings.py but it throws me an error,
Error: Failed to render ‘main.in.c’
constraints.pnot found in context while rendering ‘main.in.c’
So I also defined,
throws me another error saying ‘Code generation of ‘filename_expl_ode_fun’ is not possible since variables [p] are free’
How do I properly implement this in the toolbox?
I have nu inputs in the nonlinear model and some these inputs are set to be equal to zero at each stage. I am doing it by updating ‘lbu’ and ‘ubu’ for entire input vector u at each stage. How can I choose some entries of vector u to be set to zero over each stage?
I have certain constraints on u like constant*u1 - u2 <=0. I believe it goes in path constraint as per the problem_formulation_ocp_mex.pdf but still it is not clear which path constraint.
I have linear least square cost for state reference tracking in my formulation ||x-x_ref||^2. I understand that W and W_e are the panalties for the reference tracking. However, I don’t understant how V_x, V_u and V_z should be defined. I see that from race cars example that they are set to Identity matrices. Does it mean that by setting certain diagonal entries in V_x/V_u/V_z to zero I can define which of variables has no tracking objective?
I am sorry if some of the issues raised are trivial but wanted to be clear with the implementation I am doing. Thanks a lot in advance!