source/isaaclab_tasks/test/test_environment_determinism.py

# Copyright (c) 2022-2026, The Isaac Lab Project Developers (https://github.com/isaac-sim/IsaacLab/blob/main/CONTRIBUTORS.md).
# All rights reserved.
#
# SPDX-License-Identifier: BSD-3-Clause

"""Launch Isaac Sim Simulator first."""

from isaaclab.app import AppLauncher

# launch the simulator
app_launcher = AppLauncher(headless=True)
simulation_app = app_launcher.app


"""Rest everything follows."""

import gymnasium as gym
import pytest
import torch

import carb
import omni.usd

import isaaclab_tasks  # noqa: F401
from isaaclab_tasks.utils.parse_cfg import parse_env_cfg


@pytest.fixture(scope="module", autouse=True)
def setup_environment():
    # this flag is necessary to prevent a bug where the simulation gets stuck randomly when running the
    # test on many environments.
    carb_settings_iface = carb.settings.get_settings()
    carb_settings_iface.set_bool("/physics/cooking/ujitsoCollisionCooking", False)


@pytest.mark.parametrize(
    "task_name",
    [
        "Isaac-Open-Drawer-Franka-v0",
        "Isaac-Lift-Cube-Franka-v0",
    ],
)
@pytest.mark.parametrize("device", ["cuda", "cpu"])
def test_manipulation_env_determinism(task_name, device):
    """Check deterministic environment creation for manipulation."""
    _test_environment_determinism(task_name, device)


@pytest.mark.parametrize(
    "task_name",
    [
        "Isaac-Velocity-Flat-Anymal-C-v0",
        "Isaac-Velocity-Rough-Anymal-C-v0",
        "Isaac-Velocity-Rough-Anymal-C-Direct-v0",
    ],
)
@pytest.mark.parametrize("device", ["cuda", "cpu"])
def test_locomotion_env_determinism(task_name, device):
    """Check deterministic environment creation for locomotion."""
    _test_environment_determinism(task_name, device)


@pytest.mark.parametrize(
    "task_name",
    [
        "Isaac-Repose-Cube-Allegro-v0",
        # "Isaac-Repose-Cube-Allegro-Direct-v0",  # FIXME: @kellyg, any idea why it is not deterministic?
    ],
)
@pytest.mark.parametrize("device", ["cuda", "cpu"])
def test_dextrous_env_determinism(task_name, device):
    """Check deterministic environment creation for dextrous manipulation."""
    _test_environment_determinism(task_name, device)


def _test_environment_determinism(task_name: str, device: str):
    """Check deterministic environment creation."""
    # fix number of steps
    num_envs = 32
    num_steps = 100
    # call function to create and step the environment
    obs_1, rew_1 = _obtain_transition_tuples(task_name, num_envs, device, num_steps)
    obs_2, rew_2 = _obtain_transition_tuples(task_name, num_envs, device, num_steps)

    # check everything is as expected
    # -- rewards should be the same
    torch.testing.assert_close(rew_1, rew_2)
    # -- observations should be the same
    for key in obs_1.keys():
        torch.testing.assert_close(obs_1[key], obs_2[key])


def _obtain_transition_tuples(task_name: str, num_envs: int, device: str, num_steps: int) -> tuple[dict, torch.Tensor]:
    """Run random actions and obtain transition tuples after fixed number of steps."""
    # create a new stage
    omni.usd.get_context().new_stage()
    try:
        # parse configuration
        env_cfg = parse_env_cfg(task_name, device=device, num_envs=num_envs)
        # set seed
        env_cfg.seed = 42
        # create environment
        env = gym.make(task_name, cfg=env_cfg)
    except Exception as e:
        if "env" in locals() and hasattr(env, "_is_closed"):
            env.close()
        else:
            if hasattr(e, "obj") and hasattr(e.obj, "_is_closed"):
                e.obj.close()
        pytest.fail(f"Failed to set-up the environment for task {task_name}. Error: {e}")

    # disable control on stop
    env.unwrapped.sim._app_control_on_stop_handle = None  # type: ignore

    # reset environment
    obs, _ = env.reset()
    # simulate environment for fixed steps
    with torch.inference_mode():
        for _ in range(num_steps):
            # sample actions from -1 to 1
            actions = 2 * torch.rand(env.action_space.shape, device=env.unwrapped.device) - 1
            # apply actions and get initial observation
            obs, rewards = env.step(actions)[:2]

    # close the environment
    env.close()

    return obs, rewards