[RTR] Alignment contraint examples and bug fixes

.github/workflows/run_tests_linux.yml

@@ -36,8 +36,8 @@ jobs:
           conda list
 
       - name: Install extra dependencies
-        run: |
-          conda install pip pytest-cov black pytest pytest-cov codecov packaging pytest-mpl -cconda-forge
+        run: | 
+          conda install "numpy>=2.4,<3" pip pytest-cov black pytest pytest-cov codecov packaging pytest-mpl -cconda-forge
           sudo apt install -y librhash-dev
 
       - name: Install ACADOS on Linux

.github/workflows/run_tests_osx_win.yml

@@ -50,7 +50,7 @@ jobs:
           conda list
 
       - name: Install extra dependencies
-        run: conda install pip pytest-cov black pytest pytest-cov codecov packaging -cconda-forge
+        run: conda install "numpy>=2.4,<3" pip pytest-cov black pytest pytest-cov codecov packaging -cconda-forge
 
       - name: Install ACADOS on Mac
         run: |

.vscode/launch.json

@@ -5,51 +5,39 @@
     "version": "0.2.0",
     "configurations": [
         {
-            "name": "Run examples GUI",
+            "name": "Getting Started",
             "type": "debugpy",
             "request": "launch",
-            "module": "bioptim.examples",
+            "program": "${workspaceFolder}/bioptim/examples/getting_started/basic_ocp.py",
+            "console": "integratedTerminal",
             "justMyCode": true,
             "env": {
                 "PYTHONPATH": "${workspaceFolder}"
             },
-            "cwd": "${fileDirname}"
+            "cwd": "${workspaceFolder}/bioptim/examples/getting_started/"
         },
         {
-            "name": "Python : fichier actif",
+            "name": "Run examples GUI",
             "type": "debugpy",
             "request": "launch",
-            "program": "${file}",
-            "console": "integratedTerminal",
-            "justMyCode": false,
+            "module": "bioptim.examples",
+            "justMyCode": true,
             "env": {
                 "PYTHONPATH": "${workspaceFolder}"
             },
             "cwd": "${fileDirname}"
         },
         {
-            "name": "Pendulum",
+            "name": "Python : fichier actif",
             "type": "debugpy",
             "request": "launch",
-            "program": "${workspaceFolder}/bioptim/examples/getting_started/pendulum.py",
+            "program": "${file}",
             "console": "integratedTerminal",
-            "justMyCode": true,
+            "justMyCode": false,
             "env": {
                 "PYTHONPATH": "${workspaceFolder}"
             },
-            "cwd": "${workspaceFolder}/bioptim/examples/getting_started/"
+            "cwd": "${fileDirname}"
         },
-        {
-            "name": "Parameter",
-            "type": "debugpy",
-            "request": "launch",
-            "program": "${workspaceFolder}/bioptim/examples/muscle_driven_ocp/custom_parameters.py",
-            "console": "integratedTerminal",
-            "justMyCode": true,
-            "env": {
-                "PYTHONPATH": "${workspaceFolder}"
-            },
-            "cwd": "${workspaceFolder}/bioptim/examples/muscle_driven_ocp/"
-        }
     ]
 }

README.md

@@ -788,14 +788,15 @@ One can refer to their respective solver's documentation to know which options e
 The `show_online_optim` parameter can be set to `True` so the graphs nicely update during the optimization with the default values.
 One can also directly declare `online_optim` as an `OnlineOptim` parameter to customize the behavior of the plotter. 
 Note that `show_online_optim` and `online_optim` are mutually exclusive.
-Please also note that `OnlineOptim.MULTIPROCESS` is not available on Windows and only none of them are available on Macos. 
-To see how to run the server on Windows, please refer to the `getting_started/pendulum.py` example.
+Please also note that `OnlineOptim.MULTIPROCESS` is not available on Windows or Macos.
+On Macos, the default backend is `OnlineOptim.MULTIPROCESS_SERVER`, while `OnlineOptim.SERVER` remains available if one wants to start `resources/plotting_server.py` manually.
+To see how to run the server explicitly, please refer to the `resources/plotting_server.py` example.
 It is expected to slow down the optimization a bit. 
 `show_options` can be also passed as a dict to the plotter to customize the plotter's behavior.
 If `online_optim` is set to `SERVER`, then a server must be started manually by instantiating an `PlottingServer` class (see `ressources/plotting_server.py`).
 The following keys are additional options when using `OnlineOptim.SERVER` and `OnlineOptim.MULTIPROCESS_SERVER`:
   - `host`: the host to use (default is `localhost`)
-  - `port`: the port to use (default is `5030`)
+  - `port`: the port to use (default is `5030` for `OnlineOptim.SERVER` and a random available port for `OnlineOptim.MULTIPROCESS_SERVER`)
 
 If you want to see IPOPT's iterations over the course of the resolution of your opc, it is possible using the following:
 ```python    
@@ -1720,7 +1721,7 @@ The type of online plotter to use.
 
 The accepted values are:
 NONE: No online plotter.
-DEFAULT: Use the default online plotter depending on the OS (MULTIPROCESS on Linux, MULTIPROCESS_SERVER on Windows and NONE on MacOS).
+DEFAULT: Use the default online plotter depending on the OS (MULTIPROCESS on Linux, MULTIPROCESS_SERVER on Windows and macOS).
 MULTIPROCESS: The online plotter is in a separate process.
 SERVER: The online plotter is in a separate server.
 MULTIPROCESS_SERVER: The online plotter using the server automatically setup on a separate process.
@@ -2006,6 +2007,9 @@ available in the `biorbd` documentation.
 ### The [example_optimal_time.py](./bioptim/examples/getting_started/example_optimal_time.py) file
 Examples of time optimization can be found in 'examples/optimal_time_ocp/'.
 
+### The [example_pinocchio.py](./bioptim/examples/getting_started/example_pinocchio.py) file
+This example is the exact same as the pendulum example, but with a model defined using the `Pinocchio` backend (instead of the `biorbd` backend). It is designed to show how to use a model defined in Pinocchio instead of biorbd.
+
 ### The [example_simulation.py](./bioptim/examples/getting_started/example_simulation.py) file
 The first part of this example is a single shooting simulation from initial guesses.
 It is not an optimal control program. It is merely the simulation of values that is applying the dynamics.

bioptim/__init__.py

@@ -217,6 +217,7 @@
     JointAccelerationBiorbdModel,
     MultiTorqueBiorbdModel,
 )
+from .models.pinocchio import PinocchioModel, TorquePinocchioModel
 from .models.protocols.biomodel import BioModel
 from .models.protocols.holonomic_constraints import HolonomicConstraintsFcn, HolonomicConstraintsList
 from .models.protocols.stochastic_biomodel import StochasticBioModel

bioptim/dynamics/configure_variables.py

@@ -2,11 +2,11 @@
 from typing import Callable, Any
 
 import numpy as np
-from casadi import DM, vertcat, Function
+from casadi import DM, vertcat, Function, horzcat
 
 from .configure_new_variable import NewVariableConfiguration
 from .fatigue.fatigue_dynamics import FatigueList
-from ..misc.enums import PlotType, ContactType
+from ..misc.enums import PlotType, ContactType, ControlType
 from ..misc.fcn_enum import FcnEnum
 from ..misc.mapping import BiMapping, Mapping
 from ..models.protocols.stochastic_biomodel import StochasticBioModel
@@ -1204,20 +1204,31 @@ def configure_qv(ocp, nlp, **extra_params) -> None:
 
         time_span_sym = vertcat(nlp.time_cx, nlp.dt)
 
-        nlp.q_v_function = Function(
+        sym_qv = (
+            nlp.controls.scaled.cx
+            if nlp.control_type == ControlType.CONSTANT
+            else nlp.cx.sym("new_control", nlp.controls.scaled.cx.shape[0], 2)
+        )
+        qv_init_w_algebraic = (
+            nlp.algebraic_states["q_v"].cx
+            if "q_v" in nlp.algebraic_states.keys()
+            else DM.zeros(nlp.model.nb_dependent_joints, 1)
+        )
+
+        q_v_plot_function = Function(
             "qv_function",
             [
                 time_span_sym,
-                nlp.states.cx,
-                nlp.controls.cx,
-                nlp.parameters.cx,
-                nlp.algebraic_states.cx,
+                nlp.states.scaled.cx,
+                sym_qv,
+                nlp.parameters.scaled.cx,
+                nlp.algebraic_states.scaled.cx,
                 nlp.numerical_timeseries.cx,
             ],
             [
                 nlp.model.compute_q_v()(
                     nlp.states["q_u"].cx,
-                    DM.zeros(nlp.model.nb_dependent_joints, 1),
+                    qv_init_w_algebraic,
                 )
             ],
             ["t_span", "x", "u", "p", "a", "d"],
@@ -1239,7 +1250,7 @@ def configure_qv(ocp, nlp, **extra_params) -> None:
         )
 
         nlp.plot["q_v"] = CustomPlot(
-            lambda t0, phases_dt, node_idx, x, u, p, a, d: nlp.q_v_function(
+            lambda t0, phases_dt, node_idx, x, u, p, a, d: q_v_plot_function(
                 np.concatenate([t0, t0 + phases_dt[nlp.phase_idx]]), x, u, p, a, d
             ),
             plot_type=PlotType.INTEGRATED,
@@ -1261,12 +1272,24 @@ def configure_qdotv(ocp, nlp, **extra_params) -> None:
         """
 
         time_span_sym = vertcat(nlp.time_cx, nlp.dt)
-        nlp.q_v_function = Function(
+
+        sym_qv = (
+            nlp.controls.scaled.cx
+            if nlp.control_type == ControlType.CONSTANT
+            else nlp.cx.sym("new_control", nlp.controls.scaled.cx.shape[0], 2)
+        )
+        qv_init_w_algebraic = (
+            nlp.algebraic_states["q_v"].cx
+            if "q_v" in nlp.algebraic_states.keys()
+            else DM.zeros(nlp.model.nb_dependent_joints, 1)
+        )
+
+        qdot_v_plot_function = Function(
             "qdot_v_function",
             [
                 time_span_sym,
                 nlp.states.scaled.cx,
-                nlp.controls.scaled.cx,
+                sym_qv,
                 nlp.parameters.scaled.cx,
                 nlp.algebraic_states.scaled.cx,
                 nlp.numerical_timeseries.cx,
@@ -1275,7 +1298,7 @@ def configure_qdotv(ocp, nlp, **extra_params) -> None:
                 nlp.model._compute_qdot_v()(
                     nlp.states.scaled["q_u"].cx,
                     nlp.states.scaled["qdot_u"].cx,
-                    DM.zeros(nlp.model.nb_dependent_joints, 1),
+                    qv_init_w_algebraic,
                 )
             ],
             ["t_span", "x", "u", "p", "a", "d"],
@@ -1297,7 +1320,7 @@ def configure_qdotv(ocp, nlp, **extra_params) -> None:
         )
 
         nlp.plot["qdot_v"] = CustomPlot(
-            lambda t0, phases_dt, node_idx, x, u, p, a, d: nlp.q_v_function(
+            lambda t0, phases_dt, node_idx, x, u, p, a, d: qdot_v_plot_function(
                 np.concatenate([t0, t0 + phases_dt[nlp.phase_idx]]), x, u, p, a, d
             ),
             plot_type=PlotType.INTEGRATED,
@@ -1319,12 +1342,25 @@ def configure_lagrange_multipliers_function(ocp, nlp: NpArrayDictOptional, **ext
         """
 
         time_span_sym = vertcat(nlp.time_cx, nlp.dt)
-        nlp.lagrange_multipliers_function = Function(
+
+        sym_qv = (
+            nlp.controls.scaled.cx
+            if nlp.control_type == ControlType.CONSTANT
+            else nlp.cx.sym("new_control", nlp.controls.scaled.cx.shape[0], 2)
+        )
+        ctrl_cx = sym_qv[:, 1] if nlp.control_type == ControlType.LINEAR_CONTINUOUS else nlp.controls.scaled.cx
+        qv_init_w_algebraic = (
+            nlp.algebraic_states["q_v"].cx
+            if "q_v" in nlp.algebraic_states.keys()
+            else DM.zeros(nlp.model.nb_dependent_joints, 1)
+        )
+
+        lagrange_multipliers_plot_function = Function(
             "lagrange_multipliers_function",
             [
                 time_span_sym,
                 nlp.states.scaled.cx,
-                nlp.controls.scaled.cx,
+                sym_qv,
                 nlp.parameters.scaled.cx,
                 nlp.algebraic_states.scaled.cx,
                 nlp.numerical_timeseries.cx,
@@ -1333,8 +1369,11 @@ def configure_lagrange_multipliers_function(ocp, nlp: NpArrayDictOptional, **ext
                 nlp.model.compute_the_lagrangian_multipliers()(
                     nlp.states.scaled["q_u"].cx,
                     nlp.states.scaled["qdot_u"].cx,
-                    DM.zeros(nlp.model.nb_dependent_joints, 1),
-                    DynamicsFunctions.get(nlp.controls["tau"], nlp.controls.scaled.cx),
+                    qv_init_w_algebraic,
+                    DynamicsFunctions.get(
+                        nlp.controls["tau"],
+                        ctrl_cx,
+                    ),
                 )
             ],
             ["t_span", "x", "u", "p", "a", "d"],
@@ -1360,7 +1399,7 @@ def configure_lagrange_multipliers_function(ocp, nlp: NpArrayDictOptional, **ext
         )
 
         nlp.plot["lagrange_multipliers"] = CustomPlot(
-            lambda t0, phases_dt, node_idx, x, u, p, a, d: nlp.lagrange_multipliers_function(
+            lambda t0, phases_dt, node_idx, x, u, p, a, d: lagrange_multipliers_plot_function(
                 np.concatenate([t0, t0 + phases_dt[nlp.phase_idx]]), x, u, p, a, d
             ),
             plot_type=PlotType.INTEGRATED,

bioptim/examples/__main__.py

@@ -51,6 +51,7 @@
                     ("Example simulation", "example_simulation.py"),
                     ("Example cyclic movement", "example_cyclic_movement.py"),
                     ("Example constraint weight", "custom_constraint_weights.py"),
+                    ("Example Pinocchio", "example_pinocchio.py"),
                     ("How to plot", "how_to_plot.py"),
                 ]
             ),

bioptim/examples/getting_started/basic_ocp.py

@@ -156,8 +156,7 @@ def main():
     ocp.print(to_console=False, to_graph=False)
 
     # --- Solve the ocp --- #
-    # Default is OnlineOptim.MULTIPROCESS on Linux, OnlineOptim.MULTIPROCESS_SERVER on Windows and None on MacOS
-    # To see the graphs on MacOS, one must run the server manually (see resources/plotting_server.py)
+    # Default is OnlineOptim.MULTIPROCESS_SERVER on all platforms.
     solver = Solver.IPOPT(online_optim=OnlineOptim.DEFAULT)
 
     # # Show the constraints Jacobian sparsity
@@ -180,13 +179,16 @@ def main():
     sol.print_cost()
     # sol.graphs(show_bounds=True, save_name="results.png")
 
-    # --- Animate the solution --- #
-    viewer = "bioviz"
-    # viewer = "pyorerun"
-    sol.animate(n_frames=0, viewer=viewer, show_now=True)
+    # # --- Animate the solution --- #
+    # # To animate the solution, we can use the animate function. Uncomment the current block
+    # # to see the animation.
+    # # 'bioviz' must be installed to use the bioviz viewer.
+    # # Similarly, 'pyorerun' must be installed to use the pyorerun viewer.
+    # viewer = "bioviz"  # "pyorerun"
+    # sol.animate(n_frames=0, viewer=viewer, show_now=True)
 
     # # --- Saving the solver's output after the optimization --- #
-    # Here is an example of how we recommend to save the solution. Please note that sol.ocp is not picklable and that sol will be loaded using the current bioptim version, not the version at the time of the generation of the results.
+    # # Here is an example of how we recommend to save the solution. Please note that sol.ocp is not picklable and that sol will be loaded using the current bioptim version, not the version at the time of the generation of the results.
     # import pickle
     # import git
     # from datetime import date