Hessian of the Lagrangian

aburda · February 12, 2025, 11:17am

Hi

I want to derive some constants for an algorithm and need to take a look at the Hessian of the Lagrangian. Is there a straightforward way to get it in acados? Interface doesn’t matter.

Best,
Artyom

FreyJo · February 12, 2025, 12:46pm

Hi,

The function AcadosOcpSolver.qp_diagnostics() is a great starting point for what you want to do:

github.com

acados/acados/blob/faf2f79b73b58365d81b9e0c73e81b658076803b/interfaces/acados_template/acados_template/acados_ocp_solver.py#L1193


      
                      del solution[k]
          
              # save
              with open(filename, 'w') as f:
                  json.dump(solution, f, default=make_object_json_dumpable, indent=4, sort_keys=True)
          
              if verbose:
                  print("stored current iterate in ", os.path.join(os.getcwd(), filename))
          
          
          def qp_diagnostics(self, hessian_type: str = 'FULL_HESSIAN'):
                  """
                  Compute some diagnostic values for the last QP.
                  result = ocp_solver.qp_diagnostics(hessian_type). Possible values are
                  'FULL_HESSIAN' or 'PROJECTED_HESSIAN'
          
                  returns a dictionary with the following fields:
                  - min_eigv_stage: dict with minimum eigenvalue for each Hessian block.
                  - max_eigv_stage: dict with maximum eigenvalue for each Hessian block.
                  - condition_number_stage: dict with condition number for each Hessian block.
                  - condition_number_global: condition number for the full Hessian.

Best,
Jonathan

aburda · February 12, 2025, 1:24pm

Hi Jonathan,

That’s exactly what I was looking for, thanks!

Best,
Artyom