Tutorial 0.1: Hello World

The same as Tutorial 0.0, but way more pythonic

Imports

pyb2d is imported as b2d

import b2d
import numpy as np

These imports are only needed for plotting. b2d.plot requires OpenCV to be installed!

import b2d.plot
import matplotlib.pyplot as plt

The first step with Box2D is the creation of the world. The world is parametrized by a gravity vector.

# the world
world = b2d.world(gravity=(0, -10))

Create a circle-shaped body

# create the dynamic body
body = world.create_dynamic_body(
    position=(0, 0),
    fixtures=b2d.fixture_def(shape=b2d.circle_shape(radius=1), density=1),
)

We can now have a look at the world: We render the world st. each meter in the Box2D world will be 100 pixels in the image:

pixels_per_meter = 100
b2d.plot.plot_world(world, ppm=pixels_per_meter)
plt.show()

Lets run the world for a total of 5 seconds. Usually one wants to run the world at a certain frame rate. With the frame rate and the total time we can compute the delta for each iteration and how many steps we need

t = 5
fps = 40
dt = 1.0 / fps
n_steps = int(t / dt + 0.5)
print(f"t={t} fps={fps} dt={dt} n_steps={n_steps}")

t=5 fps=40 dt=0.025 n_steps=200

in each step we query the bodies position and velocity and store then for later plotting

positions = np.zeros([n_steps, 2])
velocites = np.zeros([n_steps, 2])
timepoints = np.zeros([n_steps])

do it

t_elapsed = 0.0
for i in range(n_steps):

    # get the bodies center of mass
    positions[i, :] = body.world_center

    # get the bodies velocity
    velocites[i, :] = body.linear_velocity

    timepoints[i] = t_elapsed

    world.step(time_step=dt, velocity_iterations=1, position_iterations=1)
    t_elapsed += dt

plot the y-position against the time. We can see that the body is falling down in an accelerating way:

plt.plot(timepoints, positions[:, 1])
plt.show()

as expected the x position is not changing since the gravity vector is non-zero only in the x direction

plt.plot(timepoints, positions[:, 0])
plt.show()