Randomizing gravity

76 · Domain randomization

Five pendulums released from 30 degrees, each in a world with its own gravity, set with model.set_gravity(g, world=w). The period shrinks as 1 / sqrt(g), matching the point-mass estimate. Steel reads model.gravity every step; the script notifies MODEL_PROPERTIES, which matters when the solver reuses carried kinematics (reuse_kinematics), where a gravity change otherwise lags by one step.

drag to orbit · scroll to zoom

0.00 s

What to look at

The weakest gravity swings slowest.
Every world swings with the same amplitude.

Key parameter

model.set_gravity(g, world=w)

Sets one world's gravity vector; model.gravity has one entry per world plus one for the global world.

Try changing

Tilt the gravity vector a few degrees per world instead of scaling it.
If your observations use the gravity vector, normalize it: it is an acceleration, not a unit vector.

Concept

When Steel sees a randomized value

What it is

Steel reads this input from the model or state every step, so a write takes effect on the next step. The tutorial still notifies where a flag exists, which keeps sleeping islands and carried kinematics consistent.

In this scene

Every world swings with the same amplitude.

Considerations

Cost: Call only, default solver, six-link chain and four boxes per world, 1024 worlds (1.12 ms per collide + step): MODEL_PROPERTIES 5 µs (3 µs captured). Flags measured one at a time; combined calls and feature validation can cost more. This notification does not 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.
Use it when: Slopes, payload tilts or other-planet tests, per world.

The script

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

dr_gravity.py
replay.py

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

"""Gravity randomization: per-world vectors with model.set_gravity, then notify MODEL_PROPERTIES."""

import math

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
LENGTH = 0.4


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


def build():
    pendulum = newton.ModelBuilder()
    link = pendulum.add_link(xform=tf(0.0, 0.0, 0.8))
    pendulum.add_shape_capsule(
        link,
        xform=tf(0.0, 0.0, -0.2),
        radius=0.01,
        half_height=0.19,
        cfg=newton.ModelBuilder.ShapeConfig(density=100.0),
    )
    pendulum.add_shape_sphere(link, xform=tf(0.0, 0.0, -LENGTH), radius=0.04)
    j = pendulum.add_joint_revolute(-1, link, axis=newton.Axis.Y, parent_xform=tf(0.0, 0.0, 0.8))
    pendulum.add_articulation([j])
    pendulum.joint_q[-1] = math.radians(30.0)
    builder = newton.ModelBuilder()
    builder.replicate(pendulum, WORLDS, spacing=(0.0, 0.3, 0.0))
    return builder.finalize()


def simulate(rec, label, g):
    model = build()
    solver = SolverSteel(model)
    for w in range(WORLDS):
        model.set_gravity((0.0, 0.0, -g[w]), world=w)
    solver.notify_model_changed(newton.ModelFlags.MODEL_PROPERTIES)
    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 _ in range(rec.frames(5.0)):
        q = np.degrees(state_0.joint_q.numpy())
        run.frame(state_0, **{f"g {v:.2f}": a for v, a in zip(g, q)})
        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(10)
    g = np.sort(rng.uniform(2.0, 15.0, WORLDS))  # m/s^2
    rec = Recorder()
    simulate(rec, "gravity sampled per world", g)
    rec.save()
    print(
        "small-angle period of a point mass, s:", np.round(2.0 * math.pi * np.sqrt(LENGTH / g), 2)
    )

# 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.