Ocp, docp, nlp models... #115

jbcaillau · 2024-06-21T08:09:19Z

@PierreMartinon still some things to be discussed in the process ocp -> docp -> sol -> plot. Check this old issue

The text was updated successfully, but these errors were encountered:

PierreMartinon · 2024-06-21T08:55:33Z

What do you mean ?

jbcaillau · 2024-06-24T21:03:10Z

Use case 1

using OptimalControl
using NLPModelsIpopt # solve (with Ipopt) is available (extension of CTDirect.jl)
using Plots # plot is available (extension of CTBase.jl)

@def ocp ...
sol = solve(ocp) # sol::OCPSolution
plot(sol)

Use case 2

using OptimalControl
using MyNlpSolver # has its own solve for an NLP (e.g., an ADNLPModel)

@def ocp ...
docp = DirectTranscription(ocp; model=:adnlp) # docp::ADNLPModel
nlp_sol = solve(docp)

Use case 2 (cont'ed)

Enforcing some traits (API...) for the solution return by an NLP solver, namely enforcing that the returned type has solution getter (returning a vector), one can then write:

using Plots # plot is available (extension of CTBase.jl)

sol = build(solution(nlp_sol), docp) # sol::OCPSolution, more or less OCPSolutionFromDOCP...
plot(sol)

PierreMartinon · 2024-06-25T17:56:00Z

@jbcaillau Regarding the Use Case 2, using current functions it would look like

@def ocp...
docp = directTranscription(ocp)
nlp_sol = solve(docp) 
# here I assume nlp_sol contains at least the vector of the primal variables at convergence, and maybe multipliers (see next message)
lots_of_stuff = parse_DOCP_solution(docp, nlp_sol)
sol = OCPSolutionFromDOCP_raw(docp, lots_of_stuff)
plot(sol)

So I would need to clean up a bit and combine the part

lots_of_stuff = parse_DOCP_solution(docp, nlp_sol)
sol = OCPSolutionFromDOCP_raw(docp, lots_of_stuff)

into something like

sol = OCPSolutionFromNLPSolution(docp, nlp_sol)

All in all, I think we're pretty close. Should not be too hard to finalize. The one point may be the exact format of nlp_sol: it's easy enough if we consider only the primal solution, but if we want multipliers too we'll need to choose a format (see message below)

PierreMartinon · 2024-06-25T18:11:03Z

More precisely, we can already split the calls as in

docp = directTranscription(ocp)
dsol = solve(docp, print_level=0, tol=1e-12)
sol1 = OCPSolutionFromDOCP(docp, dsol)

For a custom NLP solver we would typically get something a bit different from the dsol above (this one is an ipopt solution from NLPModelIpopt). We currently have the more low level subfunction actually used to build the OCP solution using only vectors from the NLP solution, mainly T,X,U,v and P. The function still requires the docp, but we have it in this use case.

And yes, I really need to combine the arguments somehow...

function OCPSolutionFromDOCP_raw(docp, T, X, U, v, P;
    objective=0, iterations=0, constraints_violation=0,
    message="No msg", stopping=nothing, success=nothing,
    sol_control_constraints=nothing, sol_state_constraints=nothing, sol_mixed_constraints=nothing, sol_variable_constraints=nothing, mult_control_constraints=nothing, mult_state_constraints=nothing, mult_mixed_constraints=nothing, mult_variable_constraints=nothing, mult_state_box_lower=nothing, 
    mult_state_box_upper=nothing, mult_control_box_lower=nothing, mult_control_box_upper=nothing, mult_variable_box_lower=nothing, mult_variable_box_upper=nothing)

To parse the NLP solution into this awful bunch of different variables and multipliers we have the following function, in which 'solution' is the NLP solution vector (primal variables). I think I could make the multipliers and constraints optional for users who are just interested in the optimal trajectory.

Note that taking the constraints and multipliers into account is a bit more involved since each solver could have its own way of handling the constraints (eg separate between boxes/ranges, linear, general nonlinear...). The current format follows ipopt, with box and nonlinear constraints.

parse_DOCP_solution(docp, solution, multipliers_constraints, multipliers_LB, multipliers_UB, constraints)

PierreMartinon · 2024-06-26T12:19:47Z

Also linked to the older #74

jbcaillau · 2024-06-26T16:37:10Z

@jbcaillau Regarding the Use Case 2, using current functions it would look like
@def ocp...
docp = directTranscription(ocp)
nlp_sol = solve(docp) 
# here I assume nlp_sol contains at least the vector of the primal variables at convergence, and maybe multipliers (see next message)
lots_of_stuff = parse_DOCP_solution(docp, nlp_sol)
sol = OCPSolutionFromDOCP_raw(docp, lots_of_stuff)
plot(sol)
So I would need to clean up a bit and combine the part [...]

@PierreMartinon Regarding what is returned by an arbitrary NLP solver, it is reasonable to expect that the user is able to access the primal vars (kind of basic trait that we want any solver to have) and, possibly, to the dual (= multipliers - if not, we just don't plot them). One way to enforce this can be:¹

using OptimalControl
using MyNLPSolver

@def ocp ...
docp = direct_transcription(ocp; model=:my_model)
nlp_sol = solve(docp) # Could be solve!(docp) depending on the solver
sol = build(docp; primal=nlp_sol.x, dual=nlp_sol.λ) # A simple version of parse_DOCP_solution
print(sol)

where build (a simplified version of OCPSolutionFromDOCP_raw, with a shorter name 🤭) only assumes primal and dual solutions are passed in the same order as what is defined in docp.

Julia Blue style recommends to limit upper camel-case convention to modules and types (and lowercase and underscores for functions) ↩

PierreMartinon · 2024-07-07T07:27:08Z

@jbcaillau @ocots Yes this is what I went for.
Currently only the primal, ie sol=ocp_solution_from_nlp(docp, solution)
but the extension to sol=ocp_solution_from_nlp(docp, solution, multipliers) is almost ready.

The part that will be a bit more involved is the range constraints, since they are handled separately and may be less standardized among NLP solvers. An overhaul will be needed here anyway, since this is the section with the gazillion variables. I should have a look at other solver interfaces before I rewrite this, for better portability.

Ps. the contest for a better name is open :D Maybe just build_solution ?

jbcaillau · 2024-07-07T07:34:36Z

@PierreMartinon well, I'd say that it's definitely the user's job to retrieve the multipliers properly. But you are right: although we know exactly the structure of what we pass, i.e. an NLP structured as

$$F(X) \rightarrow \min\\\ L_G \leq G(X) \leq U_G\\\ L \leq X \leq U$$

there might indeed be some variability in the way constraints are treated / multipliers are ordered by the NLP solver.

PierreMartinon · 2024-07-07T09:41:13Z

See control-toolbox/OptimalControl.jl#207 and #154

Have a nice weeekend !

ocots · 2024-07-26T09:07:42Z

Can we close?

jbcaillau · 2024-07-26T10:00:02Z

@ocots @PierreMartinon about that, @PierreMartinon are you done with changes in main on build_solution, get_NLP... and everything (use case = user wanting to retrieve an NLP to use his favourite solver) so that we can

make a new release of CTDirect
release OptimalControl 1.0.0

PierreMartinon · 2024-07-26T11:14:04Z

The only change in the API is that the nlp is now outside of the DOCP, ie getNLP has disappeared and direct_transcription now returns both

docp, nlp = direct_transcription(ocp)

while (re)setting the initial guess in the DOCP is now

set_initial_guess(docp, nlp, sol0)

The change is only visible when handling the NLP directly and has no impact on the standard solve calls. The aim was to make DOCP a non mutable struct with dimensions recognized as constants, trying to improve performance.

The only updates needed in OptimalControl are

update the call to the DOCP solving inside solve
update the docs for the NLP parts

(as usual, releases will need to be done together for OptimalControl and CTDirect, but the changes in CTDirect are already merged in main so we can test OptimalControl with a dev'd CTDirect)

ocots · 2024-07-26T12:08:09Z

If the api has changed the new release of CTDirect should have a different Y.

PierreMartinon · 2024-07-27T09:00:51Z

Yes, that would be 0.10.0 I guess.
I can prepare a PR for OptimalControl too.

jbcaillau · 2024-08-03T13:45:58Z

@ocots @PierreMartinon @0Yassine0 @frapac first try 🙂, Goddard problem

julia> @def ocp begin # definition of the optimal control problem

           tf ∈ R, variable
           t ∈ [ t0, tf ], time
           x = (r, v, m) ∈ R³, state
           u ∈ R, control

           x(t0) == [ r0, v0, m0 ]
           m(tf) == mf,         (1)
           0 ≤ u(t) ≤ 1
           r(t) ≥ r0
           0 ≤ v(t) ≤ vmax

           ẋ(t) == F0(x(t)) + u(t) * F1(x(t))

           r(tf) → max

       end

julia> docp = direct_transcription(ocp; grid_size=5000)
 
julia> nlp = get_nlp(docp)
ADNLPModel - Model with automatic differentiation backend ADModelBackend{
  ReverseDiffADGradient,
  ReverseDiffADHvprod,
  ForwardDiffADJprod,
  ReverseDiffADJtprod,
  SparseADJacobian,
  SparseReverseADHessian,
  ForwardDiffADGHjvprod,
}
  Problem name: Generic
   All variables: ████████████████████ 20005  All constraints: ████████████████████ 15004 
            free: ██████⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 5002              free: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0     
           lower: █████⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 5001             lower: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0     
           upper: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0                upper: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0     
         low/upp: ██████████⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 10002          low/upp: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0     
           fixed: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0                fixed: ████████████████████ 15004 
          infeas: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0               infeas: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0     
            nnzh: ( 99.97% sparsity)   55011           linear: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0     
                                                    nonlinear: ████████████████████ 15004 
                                                         nnzj: ( 99.97% sparsity)   105004

  Counters:
             obj: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0                 grad: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0                 cons: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0     
        cons_lin: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0             cons_nln: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0                 jcon: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0     
           jgrad: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0                  jac: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0              jac_lin: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0     
         jac_nln: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0                jprod: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0            jprod_lin: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0     
       jprod_nln: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0               jtprod: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0           jtprod_lin: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0     
      jtprod_nln: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0                 hess: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0                hprod: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0     
           jhess: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0               jhprod: ⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅ 0     


julia> @time ipopt(nlp)
This is Ipopt version 3.14.16, running with linear solver MUMPS 5.7.3.

Number of nonzeros in equality constraint Jacobian...:   105004
Number of nonzeros in inequality constraint Jacobian.:        0
Number of nonzeros in Lagrangian Hessian.............:    55011

Total number of variables............................:    20005
                     variables with only lower bounds:     5001
                variables with lower and upper bounds:    10002
                     variables with only upper bounds:        0
Total number of equality constraints.................:    15004
Total number of inequality constraints...............:        0
        inequality constraints with only lower bounds:        0
   inequality constraints with lower and upper bounds:        0
        inequality constraints with only upper bounds:        0

iter    objective    inf_pr   inf_du lg(mu)  ||d||  lg(rg) alpha_du alpha_pr  ls
   0 -1.0100000e+00 9.00e-01 2.00e+00  -1.0 0.00e+00    -  0.00e+00 0.00e+00   0
   1 -1.0096633e+00 8.66e-01 8.39e+02  -1.0 2.60e+00    -  3.18e-03 3.81e-02f  1
   2 -1.0107757e+00 7.47e-01 2.04e+03  -1.7 1.03e+00    -  3.22e-02 1.38e-01f  1
   3 -1.0113617e+00 5.12e-01 1.01e+03  -1.7 7.47e-01    -  1.31e-01 3.14e-01f  1
   4 -1.0093886e+00 3.01e-01 2.26e+03  -1.7 5.12e-01    -  4.02e-01 4.12e-01f  1
   5 -1.0066535e+00 3.01e-03 4.85e+03  -1.7 3.01e-01    -  4.98e-03 9.90e-01h  1
   6 -1.0079763e+00 3.01e-05 4.52e+02  -1.7 2.89e-01    -  2.33e-01 9.90e-01f  1
   7 -1.0083752e+00 4.10e-06 1.43e+04  -1.7 2.44e-01    -  7.47e-01 9.93e-01f  1
   8 -1.0083340e+00 4.45e-08 3.08e+00  -2.5 3.46e-02    -  1.00e+00 1.00e+00h  1
   9 -1.0083454e+00 8.05e-09 2.72e-01  -3.8 7.88e-03    -  1.00e+00 1.00e+00h  1
iter    objective    inf_pr   inf_du lg(mu)  ||d||  lg(rg) alpha_du alpha_pr  ls
  10 -1.0083655e+00 8.40e-10 1.38e-02  -5.7 2.11e-03    -  1.00e+00 1.00e+00h  1
  11 -1.0096160e+00 3.12e-07 3.57e+01  -8.6 5.98e-02    -  8.06e-01 1.00e+00h  1
  12 -1.0119286e+00 2.37e-06 1.16e+01  -8.6 1.60e-01    -  6.74e-01 8.14e-01h  1
  13 -1.0123971e+00 9.85e-07 1.42e+00  -8.6 1.05e-01    -  8.78e-01 6.88e-01h  1
  14 -1.0125381e+00 4.57e-07 2.58e-01  -8.6 1.26e-01    -  8.18e-01 8.14e-01h  1
  15 -1.0125630e+00 1.51e-07 3.05e-02  -8.6 1.04e-01    -  8.82e-01 8.73e-01f  1
  16 -1.0125660e+00 5.50e-08 1.02e-03  -8.6 7.60e-02    -  9.67e-01 9.69e-01f  1
  17 -1.0125660e+00 2.82e-08 1.27e-04  -8.6 7.61e-02    -  1.00e+00 5.00e-01f  2
  18 -1.0125726e+00 2.57e-08 3.54e-07  -9.0 5.19e-02    -  1.00e+00 1.00e+00h  1

Number of Iterations....: 18

                                   (scaled)                 (unscaled)
Objective...............:  -1.0125726087110427e+00   -1.0125726087110427e+00
Dual infeasibility......:   3.5366498671414553e-07    3.5366498671414553e-07
Constraint violation....:   7.6224774270272633e-09    2.5654078028569671e-08
Variable bound violation:   1.1132245658748600e-31    1.1132245658748600e-31
Complementarity.........:   1.2766492761943900e-09    1.2766492761943900e-09
Overall NLP error.......:   7.6224774270272633e-09    3.5366498671414553e-07


Number of objective function evaluations             = 20
Number of objective gradient evaluations             = 19
Number of equality constraint evaluations            = 20
Number of inequality constraint evaluations          = 0
Number of equality constraint Jacobian evaluations   = 19
Number of inequality constraint Jacobian evaluations = 0
Number of Lagrangian Hessian evaluations             = 18
Total seconds in IPOPT                               = 11.305

EXIT: Optimal Solution Found.
 11.312735 seconds (35.63 M allocations: 1.828 GiB, 19.71% gc time)
"Execution stats: first-order stationary"

julia> @time madnlp(nlp)
This is MadNLP version v0.8.4, running with umfpack

Number of nonzeros in constraint Jacobian............:   105004
Number of nonzeros in Lagrangian Hessian.............:    55011

Total number of variables............................:    20005
                     variables with only lower bounds:     5001
                variables with lower and upper bounds:    10002
                     variables with only upper bounds:        0
Total number of equality constraints.................:    15004
Total number of inequality constraints...............:        0
        inequality constraints with only lower bounds:        0
   inequality constraints with lower and upper bounds:        0
        inequality constraints with only upper bounds:        0

iter    objective    inf_pr   inf_du lg(mu)  ||d||  lg(rg) alpha_du alpha_pr  ls
   0 -1.0100000e+00 9.00e-01 2.00e+00  -1.0 0.00e+00    -  0.00e+00 0.00e+00   0
   1 -1.0096633e+00 8.66e-01 4.25e+01  -1.0 2.60e+00    -  3.18e-03 3.81e-02h  1
   2 -1.0100423e+00 7.03e-01 6.82e+01  -1.0 8.66e-01    -  4.11e-02 1.88e-01h  1
   3 -1.0083987e+00 5.14e-01 8.02e+00  -1.0 7.03e-01    -  4.71e-01 2.69e-01h  1
   4 -1.0061304e+00 5.14e-03 1.01e+02  -1.0 5.14e-01    -  2.20e-02 9.90e-01h  1
   5 -1.0081414e+00 1.52e-03 2.59e+01  -1.0 4.28e-01    -  8.87e-01 7.04e-01h  1
   6 -1.0084646e+00 1.51e-05 1.38e+00  -1.0 1.43e-01    -  1.00e+00 9.90e-01h  1
   7 -1.0085001e+00 1.91e-07 1.35e+01  -1.0 3.29e-02    -  1.00e+00 9.96e-01h  1
   8 -1.0084940e+00 6.88e-10 1.82e-03  -1.0 1.34e-02    -  1.00e+00 1.00e+00h  1
   9 -1.0084940e+00 3.33e-16 1.25e+04  -3.8 2.50e-06    -  9.99e-01 1.00e+00h  1
iter    objective    inf_pr   inf_du lg(mu)  ||d||  lg(rg) alpha_du alpha_pr  ls
  10 -1.0085019e+00 1.44e-11 3.37e-06  -3.8 7.52e-04    -  1.00e+00 1.00e+00h  1
  11 -1.0085172e+00 1.08e-10 3.94e+01  -8.6 1.59e-03    -  9.97e-01 1.00e+00h  1
  12 -1.0114077e+00 3.68e-06 1.52e+01  -8.6 2.22e-01    -  6.13e-01 8.64e-01h  1
  13 -1.0121777e+00 1.94e-06 2.69e+00  -8.6 1.35e-01    -  8.24e-01 6.12e-01h  1
  14 -1.0124634e+00 8.75e-07 3.72e-01  -8.6 1.38e-01    -  8.62e-01 7.13e-01h  1
  15 -1.0125548e+00 3.65e-07 4.57e-02  -8.6 1.21e-01    -  8.87e-01 8.54e-01h  1
  16 -1.0125648e+00 1.16e-07 6.41e-03  -8.6 9.21e-02    -  8.65e-01 8.52e-01h  1
  17 -1.0125660e+00 2.89e-08 5.57e-04  -8.6 1.08e-01    -  9.14e-01 9.11e-01h  1
  18 -1.0125660e+00 4.16e-09 6.65e-09  -8.6 8.27e-02    -  1.00e+00 1.00e+00h  1

Number of Iterations....: 18

                                   (scaled)                 (unscaled)
Objective...............:  -1.0125660371054928e+00   -1.0125660371054928e+00
Dual infeasibility......:   6.6492546317747740e-09    6.6492546317747740e-09
Constraint violation....:   4.1625690672120186e-09    4.1625690672120186e-09
Complementarity.........:   3.1425345666755113e-09    3.1425345666755113e-09
Overall NLP error.......:   6.6492546317747740e-09    6.6492546317747740e-09

Number of objective function evaluations             = 19
Number of objective gradient evaluations             = 19
Number of constraint evaluations                     = 19
Number of constraint Jacobian evaluations            = 19
Number of Lagrangian Hessian evaluations             = 18
Total wall-clock secs in solver (w/o fun. eval./lin. alg.)  =  1.733
Total wall-clock secs in linear solver                      =  4.475
Total wall-clock secs in NLP function evaluations           =  6.742
Total wall-clock secs                                       = 12.950

EXIT: Optimal Solution Found (tol = 1.0e-08).
 12.951147 seconds (35.64 M allocations: 39.980 GiB, 23.45% gc time)
"Execution stats: Optimal Solution Found (tol = 1.0e-08)."

jbcaillau · 2024-08-03T14:40:54Z

@PierreMartinon for the record: on the current release of CTDirect (v0.10) and associated syntax: to what extent do you need the docp and not the original ocp for post-processing (= plotting, if I am not missing sth)? is it just that you need the time_grid, or sth else?

jbcaillau · 2024-08-04T16:44:51Z

@PierreMartinon for the record: on the current release of CTDirect (v0.10) and associated syntax: to what extent do you need the docp and not the original ocp for post-processing (= plotting, if I am not missing sth)? is it just that you need the time_grid, or sth else?

@PierreMartinon please answer 🥹 and close!

PierreMartinon · 2024-08-07T16:53:09Z

Not sure I understand your question: you only need the solution for plot. You do need the docp to build the ocp functional solution from the discrete docp solution. For instance from https://github.com/control-toolbox/CTDirect.jl/blob/bolza/test/suite/test_misc.jl

    docp, nlp = direct_transcription(ocp)
    tag = CTDirect.MadNLPTag()
    dsol = CTDirect.solve_docp(tag, docp, nlp, display=false)
    sol = OptimalControlSolution(docp, dsol)
    sol = OptimalControlSolution(docp, primal=dsol.solution)
    sol = OptimalControlSolution(docp, primal=dsol.solution, dual=dsol.multipliers)

You can then do plot(sol) normally.

Or is it something else ?

Ps. The code above tests some internal functions for the NLP handling, normally you would just run sol = solve(ocp, :madnlp)

jbcaillau · 2024-08-07T21:53:39Z

@PierreMartinon OK. Fine! Closing.

github-actions bot assigned joseph-gergaud Jun 21, 2024

jbcaillau mentioned this issue Jun 21, 2024

Solve error with v0.7.7 control-toolbox/OptimalControl.jl#168

Closed

github-actions bot assigned PierreMartinon Jun 21, 2024

This was referenced Jun 24, 2024

OptimalControl.jl upgrade: CTBase.jl v0.10, CTFlows.jl v0.4, CTDirect.jl v0.7, CTProblems.jl v0.7 control-toolbox/OptimalControl.jl#187

Closed

WIP - ocp, docp #125

Closed

jbcaillau linked a pull request Jun 24, 2024 that will close this issue

WIP - ocp, docp #125

Closed

This was referenced Jul 7, 2024

build_solution #154

Merged

build_solution control-toolbox/OptimalControl.jl#207

Merged

jbcaillau mentioned this issue Jul 26, 2024

Release 1.0.0 control-toolbox/OptimalControl.jl#251

Open

ocots added the enhancement New feature or request label Jul 26, 2024

jbcaillau closed this as completed Aug 7, 2024

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Ocp, docp, nlp models... #115

Ocp, docp, nlp models... #115

jbcaillau commented Jun 21, 2024

PierreMartinon commented Jun 21, 2024

jbcaillau commented Jun 24, 2024 •

edited

Loading

PierreMartinon commented Jun 25, 2024 •

edited

Loading

PierreMartinon commented Jun 25, 2024 •

edited

Loading

PierreMartinon commented Jun 26, 2024 •

edited

Loading

jbcaillau commented Jun 26, 2024 •

edited

Loading

PierreMartinon commented Jul 7, 2024 •

edited

Loading

jbcaillau commented Jul 7, 2024

PierreMartinon commented Jul 7, 2024

ocots commented Jul 26, 2024

jbcaillau commented Jul 26, 2024

PierreMartinon commented Jul 26, 2024

ocots commented Jul 26, 2024

PierreMartinon commented Jul 27, 2024

jbcaillau commented Aug 3, 2024 •

edited

Loading

jbcaillau commented Aug 3, 2024 •

edited

Loading

jbcaillau commented Aug 4, 2024

PierreMartinon commented Aug 7, 2024 •

edited

Loading

jbcaillau commented Aug 7, 2024

Ocp, docp, nlp models... #115

Ocp, docp, nlp models... #115

Comments

jbcaillau commented Jun 21, 2024

PierreMartinon commented Jun 21, 2024

jbcaillau commented Jun 24, 2024 • edited Loading

Use case 1

Use case 2

Use case 2 (cont'ed)

PierreMartinon commented Jun 25, 2024 • edited Loading

PierreMartinon commented Jun 25, 2024 • edited Loading

PierreMartinon commented Jun 26, 2024 • edited Loading

jbcaillau commented Jun 26, 2024 • edited Loading

Footnotes

PierreMartinon commented Jul 7, 2024 • edited Loading

jbcaillau commented Jul 7, 2024

PierreMartinon commented Jul 7, 2024

ocots commented Jul 26, 2024

jbcaillau commented Jul 26, 2024

PierreMartinon commented Jul 26, 2024

ocots commented Jul 26, 2024

PierreMartinon commented Jul 27, 2024

jbcaillau commented Aug 3, 2024 • edited Loading

jbcaillau commented Aug 3, 2024 • edited Loading

jbcaillau commented Aug 4, 2024

PierreMartinon commented Aug 7, 2024 • edited Loading

jbcaillau commented Aug 7, 2024

jbcaillau commented Jun 24, 2024 •

edited

Loading

PierreMartinon commented Jun 25, 2024 •

edited

Loading

PierreMartinon commented Jun 25, 2024 •

edited

Loading

PierreMartinon commented Jun 26, 2024 •

edited

Loading

jbcaillau commented Jun 26, 2024 •

edited

Loading

PierreMartinon commented Jul 7, 2024 •

edited

Loading

jbcaillau commented Aug 3, 2024 •

edited

Loading

jbcaillau commented Aug 3, 2024 •

edited

Loading

PierreMartinon commented Aug 7, 2024 •

edited

Loading