Viaduct kit

Mesh and SDF contacts

A robot arm seats the top tier of a toy viaduct onto its pegs.

Play0.00 s

¼×½×1×2×

Loop

The steel preset (the default), recordedon an NVIDIA RTX PRO 6000 Blackwell.

Run it

cd examples
# the default preset (steel), then the others
uv run shelf play viaduct_kit
# change solver settings live
uv run shelf tune viaduct_kit

Source: examples/shelf/03_mesh_sdf_contacts/viaduct_kit/ · ported from steel-webgpu assets/scenes/viaduct_kit (hero_montage export) @ ad03e49

Presets

Preset	Solver	Status	Checked
steeldefault	steel	works	2026-11-10, newton f6fc82a8b, NVIDIA RTX PRO 6000 Blackwell Server Edition

Solvers with no row are untested on this example. The full settings are inpresets.toml.

Code

example.py

examples/shelf/03_mesh_sdf_contacts/viaduct_kit/example.py

# SPDX-FileCopyrightText: Copyright (c) 2026 Skild AI and the Steel contributors
# SPDX-License-Identifier: Apache-2.0
# Direct source: steel-webgpu; see presets.toml and THIRD_PARTY.md.
# Modified for Steel Shelf: preset-driven execution and shared runner integration.

###########################################################################
# Example Viaduct kit
#
# A robot arm lifts the upper tier of a toy viaduct and seats it onto the keyed pegs of the lower tier.
#
# Rebuilt from steel-webgpu's exported scene (assets/scenes/viaduct_kit: scene.json
# and scene.bin, copied next to this file); see common/exported_scene.py. The
# visual meshes draw when a steel-webgpu checkout sits next to the shelf.
#
# Command: python -m shelf run viaduct_kit
#
###########################################################################

import newton.examples
from common.exported_scene import ExportedSceneExample


class Example(ExportedSceneExample):
    file = __file__


if __name__ == "__main__":
    parser = Example.create_parser()
    viewer, args = newton.examples.init(parser)
    newton.examples.run(Example(viewer, args), args)

example.py on GitHub

presets.toml

examples/shelf/03_mesh_sdf_contacts/viaduct_kit/presets.toml

example = "viaduct_kit"
source = "steel-webgpu assets/scenes/viaduct_kit (hero_montage export) @ ad03e49"
default = "steel"

[check]
# Arguments `python -m shelf check` passes in test mode.
args = []

[gallery]
short = "A robot arm seats the top tier of a toy viaduct onto its pegs."
title = "Viaduct kit"
frame = 420

[xpbd]
status = "untested"

[vbd]
status = "untested"

[mujoco]
status = "untested"

[featherstone]
status = "untested"

[semi_implicit]
status = "untested"

[feather_pgs]
status = "untested"

[steel]
status = "works"
# 4 substeps x 8 iterations, immediate response, body-pair contact reduction: 7.1 ms a frame on an RTX PRO 6000
# (the port's colored 4 x 16: 73-93 ms, swinging run to run). The tier (124 hulls) seated on the kit (222 hulls) made
# 680-725 contacts and a dense 94 ms Steel substep; reducing contacts per body pair (Newton's default 0.25 m cells,
# as the montage was built) leaves ~125, and the tier still seats within 0.3 mm of its pegs, as in a 16 x 32 run;
# penetration 0.6 mm peak, 0.04 mm mean. 0.06 m cells were slower and deeper (0.8 / 0.29 mm). 2 and 3 substeps drop
# the tier from the grip on the way; the colored response and 16 iterations seat it too but cost 7.4-9.4 ms.
# Sleeping stays on as the port had it (not timed off). Joint limits for the fingers, built within 0.1 of a limit; no joint has a
# finite velocity limit.
sim = { fps = 60, substeps = 4 }
collision = { body_pair_reduction = {} }
solver = { solve_mode = "matrix_free", iterations = 8, articulated_contact_response = "immediate", enable_joint_limits = true, joint_limit_activation_gap = 0.1, enable_joint_velocity_limits = false, enable_sleeping = true, free_body_constraint_max = 8192, articulation_constraint_max = 1024 }
checked = "2026-11-10, newton f6fc82a8b, NVIDIA RTX PRO 6000 Blackwell Server Edition"

[kamino]
status = "untested"

[implicit_mpm]
status = "untested"

[style3d]
status = "untested"

presets.toml on GitHub

← Robot panda hydroAll examplesBasic mimic joint →