Randomizing mass

65 · Domain randomization

Five copies of a drive-held arm in five worlds. At 1 s the script scales each world's body_mass and body_inertia by a sampled factor (and the inverse arrays by its reciprocal), then calls notify_model_changed(BODY_INERTIAL_PROPERTIES). A heavier arm sags further against the same drive stiffness, roughly in proportion to its mass at these small angles. With the default update_mass_matrix_interval=1 Steel rebuilds the mass matrix from body_mass every step, but the notification also refreshes the solver's cached spatial inertias and forces a rebuild, so call it whenever you write these arrays.

drag to orbit · scroll to zoom

0.00 s

What to look at

All arms are identical for the first second.
After 1 s the sag angle grows with the sampled mass factor.

Key parameter

solver.notify_model_changed(ModelFlags.BODY_INERTIAL_PROPERTIES)

Covers body_mass, body_inv_mass, body_inertia, body_inv_inertia and body_com. Keep each inverse array consistent with its forward array.

Try changing

Scale only body_mass and leave the inertia alone, as many mass randomizers do.
Widen the range to x0.2 to x5 and lower KE until the heaviest arm touches its own hinge limit.
Write the arrays from a Warp kernel instead of NumPy so the randomization stays on the GPU.

Concept

When Steel sees a randomized value

What it is

Notify with the matching flags after changing these model arrays so Steel refreshes all derived data consistently. Some inputs are cached until notification; mass, inertia and drive gains can also be picked up by periodic mass-matrix rebuilds, so their behavior without notification depends on update_mass_matrix_interval.

In this scene

After 1 s the sag angle grows with the sampled mass factor.

Considerations

Cost: Call only, default solver, six-link chain and four boxes per world, 1024 worlds (1.12 ms per collide + step): BODY_INERTIAL_PROPERTIES 31 µs (17 µs captured). Flags measured one at a time; combined calls and feature validation can cost more. Body-inertial and joint-DOF notifications also request a mass-matrix refresh.
Graph capture: These flags can be captured in a reset graph. Kinematic-flag changes, sleeping, and shape flags with sparse factors plus restitution run eagerly; captured calls skip some value checks.
Interactions: world_mask limits which worlds lose their warm start; cached data is refreshed for every world.
Use it when: Payload and link-mass variation; scale inertia with mass unless you want a different shape.

The script

cd tutorials && python dr_mass.py prints each variant's final metric; --record out.json also writes the replay. Every tutorial imports replay.py from the same folder.

dr_mass.py
replay.py

# SPDX-FileCopyrightText: Copyright (c) 2026 Skild AI and the Steel contributors
# SPDX-License-Identifier: Apache-2.0

"""Mass randomization: per-world body mass and inertia + BODY_INERTIAL_PROPERTIES."""

import newton
import numpy as np
import warp as wp
from newton.solvers import SolverSteel

from replay import Recorder

DT = 1.0 / 240.0
SUBSTEPS = 8
WORLDS = 5
KE, KD = 15.0, 1.0  # hinge drive holding the arm level


def tf(x, y, z):
    return wp.transform((x, y, z), wp.quat_identity())


def build():
    arm = newton.ModelBuilder()
    link = arm.add_link(xform=tf(0.0, 0.0, 0.8))
    arm.add_shape_box(link, xform=tf(0.2, 0.0, 0.0), hx=0.2, hy=0.02, hz=0.02)
    arm.add_shape_sphere(link, xform=tf(0.4, 0.0, 0.0), radius=0.05)
    j = arm.add_joint_revolute(
        -1, link, axis=newton.Axis.Y, parent_xform=tf(0.0, 0.0, 0.8), target_ke=KE, target_kd=KD
    )
    arm.add_articulation([j])
    builder = newton.ModelBuilder()
    builder.replicate(arm, WORLDS, spacing=(0.0, 0.35, 0.0))
    return builder.finalize()


def randomize_mass(model, scale):
    """Scale every body of world w by scale[w]; keep the inverse arrays consistent."""
    s = scale[model.body_world.numpy()]
    model.body_mass.assign(model.body_mass.numpy() * s)
    model.body_inv_mass.assign(model.body_inv_mass.numpy() / s)
    model.body_inertia.assign(model.body_inertia.numpy() * s[:, None, None])
    model.body_inv_inertia.assign(model.body_inv_inertia.numpy() / s[:, None, None])


def simulate(rec, label, scale):
    model = build()
    solver = SolverSteel(model)
    state_0, state_1, control = model.state(), model.state(), model.control()
    newton.eval_fk(model, model.joint_q, model.joint_qd, state_0)

    run = rec.variant(label, model)
    for frame in range(rec.frames(4.0)):
        if frame == rec.fps:  # randomize at 1 s, as an episode reset would
            randomize_mass(model, scale)
            solver.notify_model_changed(newton.ModelFlags.BODY_INERTIAL_PROPERTIES)
        sag = np.degrees(state_0.joint_q.numpy())
        run.frame(state_0, **{f"mass x{s:.2f}": a for s, a in zip(scale, sag)})
        for _ in range(SUBSTEPS):
            solver.step(state_0, state_1, control, None, DT)
            state_0, state_1 = state_1, state_0
    run.check(solver)


if __name__ == "__main__":
    rng = np.random.default_rng(0)
    scale = np.sort(rng.uniform(0.5, 2.0, WORLDS))
    rec = Recorder()
    simulate(rec, "5 worlds, mass x0.5 to x2", scale)
    rec.save()

# SPDX-FileCopyrightText: Copyright (c) 2026 Skild AI and the Steel contributors
# SPDX-License-Identifier: Apache-2.0

"""Record Steel tutorial runs as replay JSON for the tutorials page, or print a summary.

Every tutorial imports this module. ``python NN_name.py`` runs the variants and prints
the final metric of each; ``--record out.json`` also writes body poses, shapes and metric
series that the page replays in 3D.
"""

import argparse
import base64
import json
import os
import subprocess
import sys
import time

import newton
import numpy as np
import warp as wp

from source_hash import source_sha256

_RENDERED_TYPES = {
    newton.GeoType.PLANE,
    newton.GeoType.SPHERE,
    newton.GeoType.CAPSULE,
    newton.GeoType.ELLIPSOID,
    newton.GeoType.CYLINDER,
    newton.GeoType.BOX,
    newton.GeoType.MESH,
    newton.GeoType.CONE,
    newton.GeoType.CONVEX_MESH,
}


class Recorder:
    """Collect the variants of one tutorial; ``fps`` is the replay frame rate."""

    def __init__(self, fps: int = 30):
        parser = argparse.ArgumentParser()
        parser.add_argument(
            "--record", metavar="OUT.json", help="write a replay file for the tutorials page"
        )
        parser.add_argument("--device", default=None, help="Warp device, for example cuda:0")
        self.args = parser.parse_args()
        if self.args.device:
            wp.set_device(self.args.device)
        self.fps = fps
        self.script = os.path.abspath(sys.argv[0])
        self.variants: list[Variant] = []

    def frames(self, seconds: float) -> int:
        """Number of replay frames in ``seconds``, including the first."""
        return int(round(seconds * self.fps)) + 1

    def variant(
        self,
        label: str,
        model: newton.Model,
        *,
        ghost: bool = False,
        slot: int | None = None,
        scene: str | None = None,
    ) -> "Variant":
        """Start one run; ``ghost`` overlays run ``slot``, ``scene`` groups runs for charts."""
        variant = Variant(label, model, self.fps, ghost, slot, scene)
        self.variants.append(variant)
        return variant

    def save(self) -> None:
        """Print a summary and, with ``--record``, write the replay file."""
        for variant in self.variants:
            last = variant.metrics[-1] if variant.metrics else {}
            flag = "  (rows dropped)" if variant.overflow else ""
            print(f"{variant.label:<48} {json.dumps(last)}  {variant.wall:.2f} s{flag}")
        if not self.args.record:
            return
        source_sha = source_sha256(self.script)
        doc = {
            "script": os.path.basename(self.script),
            "script_sha256": source_sha,
            "newton_commit": _newton_commit(),
            "newton_version": newton.__version__,
            "warp_version": wp.config.version,
            "gpu": wp.get_device().name,
            "fps": self.fps,
            "variants": [v.to_json() for v in self.variants],
        }
        with open(self.args.record, "w") as f:
            json.dump(doc, f, separators=(",", ":"))
        print(f"wrote {self.args.record}")


class Variant:
    """One recorded run: a pose per frame for every body, plus named metric values."""

    def __init__(self, label, model, fps, ghost, slot, scene):
        self.label, self.model, self.fps, self.ghost, self.slot = label, model, fps, ghost, slot
        self.scene = scene
        self.poses: list[np.ndarray] = []
        self.metrics: list[dict] = []
        self.awake: list[list[int]] = []
        self.particles: list[np.ndarray] = []
        self.overflow = False
        self.start = time.perf_counter()
        self.wall = 0.0

    def frame(self, state: newton.State, awake=None, particles=None, **metrics) -> None:
        """Record ``state.body_q``, optional awake flags, particle positions and metric values."""
        self.poses.append(state.body_q.numpy().astype(np.float32).copy())
        if particles is not None:
            self.particles.append(particles.numpy().astype(np.float32).copy())
        self.metrics.append({k: _round(v) for k, v in metrics.items()})
        if awake is not None:
            self.awake.append([int(x) for x in awake.numpy()])
        self.wall = time.perf_counter() - self.start

    def check(self, solver) -> None:
        """Note whether the solver dropped rows; call once after the run."""
        try:
            solver.check_constraint_capacity()
        except RuntimeError:
            self.overflow = True

    def to_json(self) -> dict:
        poses = (
            np.stack(self.poses)
            if self.poses
            else np.zeros((0, self.model.body_count, 7), np.float32)
        )
        return {
            "label": self.label,
            "fps": self.fps,
            "ghost": self.ghost,
            "slot": self.slot,
            "scene": self.scene,
            "bodies": int(self.model.body_count),
            "frames": int(poses.shape[0]),
            "poses": base64.b64encode(poses.astype(np.float32).tobytes()).decode(),
            "shapes": _shape_table(self.model),
            "metric": self.metrics,
            "awake": self.awake or None,
            "particles": self._particles_json(),
            "wall_s": round(self.wall, 3),
            "overflow": self.overflow,
        }

    def _particles_json(self):
        if not self.particles:
            return None
        frames = np.stack(self.particles)
        tris = self.model.tri_indices
        radius = self.model.particle_radius.numpy()
        return {
            "count": int(frames.shape[1]),
            "frames": base64.b64encode(frames.astype(np.float32).tobytes()).decode(),
            "radius": round(float(radius.mean()), 5) if radius.size else 0.0,
            "tris": [int(v) for v in tris.numpy().reshape(-1)]
            if tris is not None and tris.shape[0]
            else None,
        }


def _shape_table(model):
    types = model.shape_type.numpy()
    scales = model.shape_scale.numpy()
    xforms = model.shape_transform.numpy()
    bodies = model.shape_body.numpy()
    flags = model.shape_flags.numpy()
    shapes = []
    for i, t in enumerate(types):
        if int(t) not in _RENDERED_TYPES or not flags[i] & int(newton.ShapeFlags.VISIBLE):
            continue
        shape = {
            "b": int(bodies[i]),
            "t": int(t),
            "s": [round(float(v), 5) for v in scales[i]],
            "x": [round(float(v), 6) for v in xforms[i]],
        }
        if int(t) in (newton.GeoType.MESH, newton.GeoType.CONVEX_MESH):
            mesh = model.shape_source[i]
            shape["v"] = [round(float(v), 5) for v in np.asarray(mesh.vertices).reshape(-1)]
            shape["i"] = [int(v) for v in np.asarray(mesh.indices).reshape(-1)]
        shapes.append(shape)
    return shapes


def _round(value):
    value = float(value)
    return round(value, 5) if np.isfinite(value) else None


def _newton_commit():
    if os.environ.get("NEWTON_COMMIT"):
        return os.environ["NEWTON_COMMIT"]
    try:
        root = os.path.dirname(os.path.dirname(newton.__file__))
        return subprocess.check_output(["git", "-C", root, "rev-parse", "HEAD"], text=True).strip()
    except (OSError, subprocess.CalledProcessError):
        return None

Recorded with SolverSteel on Newton 1.9.0.dev0, Warp 1.19.0, NVIDIA RTX PRO 6000 Blackwell Server Edition. Replays read the state back every frame, so they are not a speed benchmark. Script · page text

Steel docs and tutorials are licensed under Apache-2.0. Third-party notices. "Skild", "Skild AI" and the Skild logo are trademarks of Skild AI and are not licensed under Apache-2.0.