Pyomeca examples

A colleciton of simple examples codes for simpler starting with pyomeca software, particularly with biorbd and bioviz

This is an example tutorial of how to setup biorbd and bioviz with pycapacity to calculate and visualise the human capacities

Installing Pyomeca biorbd and bioviz

the best way to download and install the pyomeca biordb and bioviz libraries is to use anaconda.

For anaconda install you can simply download the yaml file and save it as env.yaml:

name: pyomeca
channels:
  - conda-forge
  - default
dependencies:
  - python=3.9
  - bioviz
  - biorbd
  - jupyter
  - pip 
  - pip:
    - pycapacity

And create a new ready to go environment:

conda env create -f env.yaml    # create the new environemnt and install biorbd,bioviz 
conda activate pyomeca

Creating the custom environment from scratch

You can also simply use anaconda to create a new custom environment:

conda create -n pyomeca python=3 # create python 3 based environment
conda activate pyomeca           # activate the environment 
conda install biorbd bioviz      # install biorbd and bioviz

Then install pycapacity for the workspace analysis

pip install pycapacity

Downloading musculoskeletal models

you can use any pyomeca .bioMod model, however to jump-start the development we suggest you to download our small database of models at the link

You can download it using terminal:

git clone https://gitlab.inria.fr/auctus-team/components/modelisation/humanmodels/pyomeca_models.git

Or downloading the zip from the link https://gitlab.inria.fr/auctus-team/components/modelisation/humanmodels/pyomeca_models

Make sure to place the pyomeca_models folder in the directory that will contain your python code.

Visualise models code example

A simple example of a python code visualising the mucsuloskeletal models. by launching this code you should be able to make sure that your installation works well.

import bioviz
import biorbd
import numpy as np

# Load a predefined model
# model = biorbd.Model("pyomeca_models/MOBL_ARMS_fixed_33.bioMod")
# model = biorbd.Model("pyomeca_models/BrasComplet.bioMod")
# model = biorbd.Model("pyomeca_models/BrasViolon.bioMod")
# model = biorbd.Model("pyomeca_models/arm26.bioMod")
model = biorbd.Model("pyomeca_models/BrasCompletIMUS.bioMod")

# get some random data
nq = model.nbQ()
nb_mus = model.nbMuscles()
print("dof number: ",nq)
print("muscle number: ",nb_mus)


# a bit of animation
q_recons = np.zeros((model.nbQ(), 100))
for i in range(1,100):
    # increment all joint values by -0.003
    q_recons[:, i] = q_recons[:, i-1]-0.003 

# double duration of the animation
q_recons = np.hstack((q_recons,np.flip(q_recons)))

# visualize
b = bioviz.Viz(loaded_model=model)
# b.load_movement(q_recons)
b.exec()

Interactive force polytope code example

A simple example of a python code visualising the mucsuloskeletal models. by launching this code you should be able to make sure that your installation works well.

import numpy as np
import biorbd
import bioviz
from bioviz.biorbd_vtk import VtkModel, VtkWindow, Mesh

# a bit of statistics
import time

# polytope algorithm
from pycapacity.human import force_polytope

# Load a predefined model
model = biorbd.Model("pyomeca_models/MOBL_ARMS_fixed_33.bioMod")

# get the number of dof and muslces
nq = model.nbQ()
nb_mus = model.nbMuscles()

# Animate the results if biorbd viz is installed
b = bioviz.Viz(loaded_model=model,
               background_color=(1,1,1), 
               show_local_ref_frame=False, 
               show_global_ref_frame=False, 
               show_markers=True,
               show_global_center_of_mass=False,
               show_segments_center_of_mass=False, 
               show_wrappings=False, 
               show_floor=False, 
               show_gravity_vector=False)
# define the meshes for the polytope - without robot
vtkMeshView = VtkModel(b.vtk_window, patch_color=[[0,0.5,0.8]],mesh_opacity=0.5)
vtkMeshView1 = VtkModel(b.vtk_window, patch_color=[[0,0.5,0.8]],mesh_opacity=0.8, force_wireframe=True)

b.set_q([0.0,1.4237,-1.256,1.8218,0.0,0.0,0.0])
while b.vtk_window.is_active:
    Q = b.Q
    model.updateMuscles(Q, True)
    model.UpdateKinematicsCustom(Q, np.zeros(nq), np.zeros(nq))

    F_max = []
    F_min = []
    for i in range(nb_mus):
        F_max.append(model.muscle(i).characteristics().forceIsoMax())
        #F_min.append(0)
        a = biorbd.HillThelenType(model.muscle(i)).FlPE()
        if a > 1:
            a = 0.1
        elif a < 0:
            a = 0
        F_min.append(a*F_max[-1])

    start = time.time()
    N = -model.musclesLengthJacobian(Q).to_array().T
    J = model.markersJacobian(Q, False, False)[-1].to_array()
    print("matrices time", time.time() - start)

    # Proceed with the inverse dynamics
    Tau_grav = model.InverseDynamics(Q, np.zeros(nq), np.zeros(nq))
    
    start = time.time()
    f_poly = force_polytope(J, N, F_min, F_max, 10, -Tau_grav.to_array())
    print("polytope time", time.time() - start)

    ## display polytope in the bioviz
    f_vert_show = f_poly.vertices/2000 + model.markers(Q)[model.nbMarkers()-1].to_array()[:,None]

    # plot polytope (blue) - with the edges
    meshes = []
    meshes.append(Mesh(vertex=f_vert_show[:,:,None], triangles=f_poly.face_indices.T))
    vtkMeshView.new_mesh_set(meshes)
    vtkMeshView1.new_mesh_set(meshes)

    # update visualisation
    b.update()

📢 NEW Examples!

For some more examples check out the examples folder of the repository.

• Interactive jupyter notebooks are available in the examples/notebooks folder: see on Github

• Python scripts are available in the examples/scripts folder: see on Github