Quick Start Example with VSF Simulation

Here we show a minimal example to visualize the simulation process of a point volumetric stiffness field (Point VSF) model.

The example demonstrates how to:

• build a simple Klampt world containing a rigid cube,

• attach a force/torque sensor to that cube,

• create a uniformly-stiff PointVSF represented by randomly sampled points,

• run a quasi-static VSF simulator while moving the cube downward, and

• stream measurements while visualizing both the world and the sensed interaction forces.

Animated preview

Copy each section below into one file (e.g. minimal_vsf_quick_start.py) in order, then run it with python minimal_vsf_quick_start.py. Splitting the code into thematic blocks makes it easier to follow and reuse.

Imports

import numpy as np
import time

from klampt import vis
from klampt.vis import Appearance
from klampt.model import create
from klampt.io import open3d_convert

from vsf import PointVSF
from vsf.sim.constructors import QuasistaticVSFSimulator
from vsf.sim.klampt_world_wrapper import klamptWorldWrapper
from vsf.visualize.klampt_visualization import vsf_to_point_cloud
from vsf.sensor.force_torque_sensor import ForceTorqueSensor

from vsf.visualize.o3d_visualization import create_vector_arrow

World and Sensor Setup

# 1. Create a Klampt world and add a rigid cube
world = klamptWorldWrapper()
box = create.box(0.1, 0.1, 0.1, center=[0.0, 0.0, 0.2])
world.add_geometry("cube", box, "rigid")
world.setup_local_pcd_lst()
world.setup_local_sdf_lst()

# 2. Attach a force/torque sensor to the cube
sensor = ForceTorqueSensor("force_torque_sensor", "cube")

VSF Creation and Simulator Initialization

# 3. Create a Point VSF: 100 k points in a 20 cm cube
vsf = PointVSF(rest_points=np.random.rand(100_000, 3) * 0.2 - 0.1)
vsf.stiffness += 1.0  # uniform stiffness

# 4. Set up the quasi-static simulator
simulator = QuasistaticVSFSimulator(world, [sensor])
simulator.add_deformable("vsf", vsf)

Visualization Setup

# 5. Visualization
vis.add("world", world.world, hideLabel=True)
vis.add("vsf", vsf_to_point_cloud(vsf, auto_stiffness_threshold=0.0), hideLabel=True)
vis.hideLabel("vsf")
vis.show()

input("Press <Enter> to start the simulation …")

Simulation Loop

# 6. Move the cube downward and visualize forces
for z in np.linspace(0.0, -0.2, 120):

    vis.lock()

    # Rigid-body control: translation in z
    control = {
        "cube": np.array(
            [
                [1.0, 0.0, 0.0, 0.0],
                [0.0, 1.0, 0.0, 0.0],
                [0.0, 0.0, 1.0, z],
                [0.0, 0.0, 0.0, 1.0],
            ]
        )
    }

    simulator.step(control, 0.1)
    vis.unlock()

    measurements = simulator.measurements()
    print("force and torque:", repr(measurements["force_torque_sensor"]))

    time.sleep(0.1)

    # 7. Draw an arrow indicating the contact force
    center = np.array([0.0, 0.0, 0.2 + z])
    end = center + measurements["force_torque_sensor"][:3].numpy() * 5e-4

    if np.linalg.norm(end - center) > 0.01:
        arrow = create_vector_arrow(end, center, scale=5.0)

        arrow_appearance = Appearance()
        arrow_appearance.setColor(1.0, 0.0, 0.0, 0.5)
        arrow_appearance.setSilhouette(0)
        arrow_appearance.setCreaseAngle(0)

        vis.add(
            "arrow",
            open3d_convert.from_open3d(arrow),
            appearance=arrow_appearance,
            hideLabel=True,
            draw_order=-9999,
        )
        vis.hideLabel("arrow")