Merge pull request #75 from control-toolbox/74-dev-up-compat

up compat ctbase

Author: ocots

Project.toml

@@ -15,7 +15,7 @@ JuMPModels = "JuMP"
 OptimalControlModels = "OptimalControl"
 
 [compat]
-CTBase = "0.15"
+CTBase = "0.16"
 JuMP = "1.25"
 OptimalControl = "1.0"
 julia = "1.10"

docs/src/list_of_problems.md

@@ -37,8 +37,8 @@ The table below summarizes the names and status of the each problem:
 | glider                        | ✅        | ❌    |
 | insurance                     | ✅        | ✅    |
 | jackson                       | ✅        | ✅    |
-| moonlander                    | ✅        | 🟠    |
-| quadrotor                     | 🟠        | 🟠    |
+| moonlander                    | ✅        | ✅    |
+| quadrotor                     | 🟠        | ✅    |
 | robbins                       | ✅        | ✅    |
 | robot                         | ✅        | ✅    |
 | rocket                        | ✅        | ✅    |
@@ -54,10 +54,10 @@ The problems are solved with Ipopt and the parameters:
 ```julia
 tol = 1e-8
 constr_viol_tol = 1e-6
-max_iter = 500
+max_iter = 1000
 mu_strategy = "adaptive"
 linear_solver = "mumps"
-max_wall_time = 240.0
+max_wall_time = 500
 sb = "yes"
 ```
 The symbols in the table means:

test/Project.toml

@@ -9,7 +9,7 @@ Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
 
 [compat]
 Aqua = "0.8"
-CTBase = "0.15"
+CTBase = "0.16"
 Ipopt = "1.9"
 JuMP = "1.25"
 NLPModelsIpopt = "0.10"

test/runtests.jl

@@ -7,6 +7,14 @@ using OptimalControl
 using OptimalControlProblems
 using Test
 
+# Parameters
+const tol = 1e-6
+const mu_strategy = "adaptive"
+const sb = "yes"
+const constr_viol_tol = 1e-6
+const max_iter = 1000
+const max_wall_time = 500.0
+
 #
 @testset verbose = true showtiming = true "OptimalControlProblems tests" begin
     for name in (

test/test_JuMP.jl

@@ -11,26 +11,28 @@ function test_JuMP()
         all_names,
     )
 
-    pbs_with_issues = [:cart_pendulum]
+    pbs_with_issues = [:cart_pendulum, :truck_trailer]
     functions_list = setdiff(functions_list, pbs_with_issues)
 
     for f in functions_list
         @testset "$(f)" begin
-            println("  $f")
+            println("  $f:")
             # Set up the model
             model = OptimalControlProblems.eval(f)(JuMPBackend())
             set_optimizer(model, Ipopt.Optimizer)
             set_silent(model)
-            set_optimizer_attribute(model, "tol", 1e-8)
-            set_optimizer_attribute(model, "constr_viol_tol", 1e-6)
-            set_optimizer_attribute(model, "max_iter", 500)
-            set_optimizer_attribute(model, "mu_strategy", "adaptive")
+            set_optimizer_attribute(model, "tol", tol)
+            set_optimizer_attribute(model, "constr_viol_tol", constr_viol_tol)
+            set_optimizer_attribute(model, "max_iter", max_iter)
+            set_optimizer_attribute(model, "mu_strategy", mu_strategy)
             set_optimizer_attribute(model, "linear_solver", "mumps")
-            set_optimizer_attribute(model, "max_wall_time", 240.0)
-            set_optimizer_attribute(model, "sb", "yes")
+            set_optimizer_attribute(model, "max_wall_time", max_wall_time)
+            set_optimizer_attribute(model, "sb", sb)
             # Solve the model
-            optimize!(model)
+            print("  First solve:  "); @time optimize!(model)
+            print("  Second solve: "); @time optimize!(model)
             # Test that the solver found an optimal solution
+            println("  termination_status = $(termination_status(model))\n")
             if f == :truck_trailer || f == :quadrotor
                 @test (termination_status(model) == MOI.LOCALLY_INFEASIBLE) || (termination_status(model) == MOI.ITERATION_LIMIT)
             else

test/test_OptimalControl.jl

@@ -11,28 +11,29 @@ function test_OptimalControl()
         all_names,
     )
 
-    pbs_with_issues = [:glider]
+    pbs_with_issues = [:glider, :moonlander]
     functions_list = setdiff(functions_list, pbs_with_issues)
 
+    kwargs = Dict(
+        :print_level => 0,
+        :tol => tol,
+        :mu_strategy => mu_strategy,
+        :sb => sb,
+        :constr_viol_tol => constr_viol_tol,
+        :max_iter => max_iter,
+        :max_wall_time => max_wall_time,
+    )
+
     for f in functions_list
-        println("  $f")
+        println("  $f:")
         @testset "$(f)" begin
             # Set up the model
             _, model = OptimalControlProblems.eval(f)(OptimalControlBackend())
-            sol = NLPModelsIpopt.ipopt(
-                model;
-                print_level=0,
-                tol=1e-8,
-                mu_strategy="adaptive",
-                sb="yes",
-                constr_viol_tol=1e-6,
-                max_iter=500,
-                max_wall_time=240.0,
-            )
+            print("  First solve:  "); @time sol = NLPModelsIpopt.ipopt(model; kwargs...)
+            print("  Second solve: "); @time sol = NLPModelsIpopt.ipopt(model; kwargs...)
+            println("  sol.status = $(sol.status)\n")
             # Test that the solver found an optimal solution
-            if  f == :moonlander    ||
-                f == :truck_trailer ||
-                f == :quadrotor     ||
+            if  f == :truck_trailer ||
                 f == :space_shuttle
                 @test (sol.status == :infeasible) || (sol.status == :max_iter)
             else