cosmic-warp 0.1.0__tar.gz

cosmic_warp-0.1.0/CHANGELOG.md

@@ -0,0 +1,49 @@
+# Changelog
+
+All notable changes to this project are documented here. The format follows
+[Keep a Changelog](https://keepachangelog.com/en/1.1.0/), and the project
+uses [semantic versioning](https://semver.org/).
+
+## [0.1.0] - 2026-11-04
+
+First public release, as a **beta**: the public API may change before 1.0,
+and each breaking change will be listed here.
+
+### Features
+
+- GPU-parallel spacecraft dynamics on NVIDIA Warp: thousands to millions of
+  independent environments, 6-DOF orbital and attitude dynamics in float64,
+  with physics, integration and termination fused into one kernel per step
+  and replayed as a CUDA graph by `simulate()`.
+- Integrators: Euler, RK4 (default) and symplectic kick–drift–kick, each
+  verified to reach its nominal order of convergence.
+- Physics: point-mass gravity, J2, atmospheric drag, gravity-gradient and
+  magnetic torques, variable mass.
+- Actuators: up to 8 thrusters (Isp, propellant use) and 4 reaction wheels
+  (momentum saturation, viscous friction) per body.
+- Multi-body environments, configured per body from Python or YAML.
+- Reset sampling (Cartesian boxes or orbital elements) and domain
+  randomization on the GPU, from counter-based (Philox) per-entity random
+  streams with 64-bit seeds; on-device termination, including non-finite
+  states and actuator commands.
+- Differentiable rollouts: reverse-mode gradients w.r.t. initial state,
+  control sequences and vehicle parameters.
+- Reinforcement learning: on-device vectorized environment with auto-reset
+  and zero-copy torch views; adapters for rsl_rl, skrl and rl_games; a
+  differentiable environment for first-order methods. A reference attitude
+  task and a minimal SHAC trainer are in `examples/`.
+- Seventeen examples, a YAML scenario format, a physics validation suite
+  against analytic results, and a per-GPU performance regression benchmark.
+
+### Known limitations
+
+- No user force/torque input and no supported custom-integrator interface;
+  both are planned for 1.x.
+- Earth is the only central body: orbit specs take no `mu`.
+- The integrators do not locate events: reaction-wheel saturation inside a
+  step limits every integrator to second order around it, and propellant
+  burnout is spread over its step (delta-v exact, position error
+  O(dt²) once per burnout).
+- On consumer GPUs the step is bound by FP64 throughput (1/64 of FP32 rate);
+  GPUs with full-rate FP64 have not been measured.
+- The RL trainer adapters support one body per environment.

cosmic_warp-0.1.0/CITATION.cff

@@ -0,0 +1,21 @@
+cff-version: 1.2.0
+message: "If you use this software, please cite it as below."
+title: "COSMIC WARP: GPU-Accelerated Spacecraft Simulator"
+type: software
+authors:
+  - family-names: "Lima"
+    given-names: "Pedro"
+    affiliation: "University of Luxembourg, SnT, SpaceR"
+  - family-names: "Arora"
+    given-names: "Aman"
+    affiliation: "University of Luxembourg, SnT, SpaceR"
+version: 0.1.0
+license: Apache-2.0
+repository-code: "https://github.com/Pedro-g-Lima/COSMIC_WARP"
+url: "https://pedro-g-lima.github.io/COSMIC_WARP/"
+keywords:
+  - spacecraft simulation
+  - GPU
+  - NVIDIA Warp
+  - differentiable simulation
+  - reinforcement learning

cosmic_warp-0.1.0/CONTRIBUTING.md

@@ -0,0 +1,83 @@
+# Contributing to COSMIC WARP
+
+Thanks for your interest! Bug reports, fixes, documentation improvements and
+new physics or actuator models are all welcome. For larger changes, open an
+issue first so the approach can be discussed before you invest time.
+
+## Development install
+
+```bash
+git clone https://github.com/Pedro-g-Lima/COSMIC_WARP.git
+cd COSMIC_WARP
+python3 -m venv .venv && source .venv/bin/activate
+pip install --upgrade "pip>=25.1"   # for --group
+pip install -e . --group dev
+# the RL stack: pip install -e ".[torch,rsl-rl,skrl,rl-games]" --group dev
+```
+
+## Tests
+
+```bash
+pytest                   # full suite on CUDA
+pytest --device cpu      # without a GPU (what CI runs)
+pytest -m "not slow"     # skip the end-to-end training tests
+```
+
+CI runs the CPU suite on Python 3.12 and 3.14; both ends of the dependency
+ranges (Python 3.12 with the lowest supported Warp, NumPy and PyYAML, Python
+3.14 with the newest); the built wheel in a fresh environment; and the RL
+tests with CPU torch. CUDA-only tests (graph capture, training, the GPU
+golden trajectories) must pass on a GPU before a release.
+
+If you change a step kernel, also run the performance gate on your GPU:
+
+```bash
+python benchmarks/bench.py --save   # once, on the base commit
+python benchmarks/bench.py          # after your change; fails on >15 % regression
+```
+
+## Releasing
+
+Set `__version__` in `cosmic_warp/__init__.py` (the package version is read
+from it) and in `CITATION.cff`, date the `CHANGELOG.md` entry, then push the
+tag `v<version>`: the release workflow builds, checks and publishes to PyPI,
+and refuses a tag that does not match `__version__`.
+
+## Docs
+
+```bash
+pip install -e . --group docs
+mkdocs build --strict    # must pass
+mkdocs serve             # preview at http://127.0.0.1:8000
+```
+
+## What a pull request needs
+
+- **Correctness first.** Physical accuracy, then differentiability, then
+  GPU performance. No silent approximations: state any model limit in the
+  docs.
+- **A test for every fix**, one that fails before the fix. Physics changes
+  are tested against an analytic or known result, and gradients against
+  finite differences. Don't loosen tolerances to make a test pass; seed
+  randomness; every skip states its reason.
+- **Validate at configuration time**, with a `ValueError` that names the
+  field. Nothing in the step loop may synchronize with the host.
+- **Float64 and adjoint-safe kernels**: see the
+  [kernel conventions](https://pedro-g-lima.github.io/COSMIC_WARP/guides/extending/#kernel-conventions).
+  Every float literal in a kernel goes through `wp.float64(...)`
+  (`tests/test_kernel_hygiene.py` enforces it).
+- **Golden trajectories.** `tests/test_golden.py` pins a seeded run bit for
+  bit. If your change is meant to alter results, regenerate the fixtures
+  (`pytest tests/test_golden.py --device cpu --update-goldens`, and on a GPU
+  with `--device cuda`) and say why in the pull request.
+- **Scope.** The package is the simulator: controllers, tasks and training
+  algorithms belong in `examples/` or in your own project.
+- **Docs and changelog.** Update the docs for any behaviour change and add
+  an entry to `CHANGELOG.md`.
+- Match the surrounding code style; keep changes minimal and focused.
+
+## License
+
+Unless you explicitly state otherwise, any contribution you submit for
+inclusion is licensed under the Apache License 2.0, as defined in section 5
+of the [license](LICENSE), without additional terms or conditions.

cosmic_warp-0.1.0/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2026 Pedro Lima and Aman Arora
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

cosmic_warp-0.1.0/MANIFEST.in

@@ -0,0 +1,10 @@
+# The sdist is the whole source (conda-forge builds and tests from it): the
+# test suite with its helpers and golden fixtures, and what the suite reads —
+# the docs it checks, the examples it runs, the scripts that draw the figures.
+graft tests
+graft docs
+graft examples
+graft scripts
+graft benchmarks
+include mkdocs.yml CHANGELOG.md CITATION.cff CONTRIBUTING.md
+global-exclude __pycache__ *.py[cod]

cosmic_warp-0.1.0/PKG-INFO

@@ -0,0 +1,251 @@
+Metadata-Version: 2.4
+Name: cosmic-warp
+Version: 0.1.0
+Summary: GPU-accelerated parallel spacecraft dynamics simulator built on NVIDIA Warp
+Author: Pedro Lima, Aman Arora
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://github.com/Pedro-g-Lima/COSMIC_WARP
+Project-URL: Documentation, https://pedro-g-lima.github.io/COSMIC_WARP/
+Project-URL: Issues, https://github.com/Pedro-g-Lima/COSMIC_WARP/issues
+Project-URL: Changelog, https://github.com/Pedro-g-Lima/COSMIC_WARP/blob/main/CHANGELOG.md
+Keywords: spacecraft,simulation,astrodynamics,attitude dynamics,gpu,nvidia warp,differentiable simulation,reinforcement learning
+Classifier: Development Status :: 4 - Beta
+Classifier: Environment :: GPU :: NVIDIA CUDA
+Classifier: Intended Audience :: Science/Research
+Classifier: Operating System :: POSIX :: Linux
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Topic :: Scientific/Engineering
+Classifier: Topic :: Scientific/Engineering :: Physics
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.12
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: warp-lang<1.19,>=1.15
+Requires-Dist: numpy>=2.0
+Requires-Dist: PyYAML>=6.0.1
+Provides-Extra: examples
+Requires-Dist: matplotlib; extra == "examples"
+Provides-Extra: torch
+Requires-Dist: torch; extra == "torch"
+Provides-Extra: rsl-rl
+Requires-Dist: rsl-rl-lib<6.0,>=5.0; extra == "rsl-rl"
+Provides-Extra: skrl
+Requires-Dist: skrl[torch]<3.0,>=2.0; extra == "skrl"
+Provides-Extra: rl-games
+Requires-Dist: rl-games<2.0,>=1.6.5; extra == "rl-games"
+Dynamic: license-file
+
+# COSMIC WARP
+
+<p align="center"><img src="https://raw.githubusercontent.com/Pedro-g-Lima/COSMIC_WARP/main/docs/assets/banner.png" alt="COSMIC WARP" width="900" /></p>
+
+<p align="center">
+<a href="https://github.com/Pedro-g-Lima/COSMIC_WARP/actions/workflows/tests.yml"><img src="https://github.com/Pedro-g-Lima/COSMIC_WARP/actions/workflows/tests.yml/badge.svg" alt="Tests" /></a>
+<a href="https://pedro-g-lima.github.io/COSMIC_WARP/"><img src="https://img.shields.io/badge/docs-GitHub%20Pages-blue" alt="Documentation" /></a>
+<a href="https://pypi.org/project/cosmic-warp/"><img src="https://img.shields.io/pypi/v/cosmic-warp" alt="PyPI" /></a>
+<img src="https://img.shields.io/badge/python-3.12%20%7C%203.13%20%7C%203.14-blue" alt="Python 3.12-3.14" />
+<a href="https://github.com/Pedro-g-Lima/COSMIC_WARP/blob/main/LICENSE"><img src="https://img.shields.io/badge/license-Apache--2.0-green" alt="License: Apache-2.0" /></a>
+</p>
+
+COSMIC WARP is a GPU-parallel spacecraft dynamics simulator built on
+[NVIDIA Warp](https://github.com/NVIDIA/warp). It propagates thousands to
+millions of independent spacecraft environments at once — 6-DOF orbital and
+attitude dynamics in float64, with thrusters, reaction wheels and
+environmental perturbations — for Monte Carlo studies, trajectory and
+parameter optimization with exact gradients, and reinforcement learning.
+`simulate()` and the RL environment replay each step as a CUDA graph, with
+no host round-trip inside the loop.
+
+**[Documentation](https://pedro-g-lima.github.io/COSMIC_WARP/)** — installation, guides,
+examples, theory and API reference.
+
+> **Status: 0.1 beta.** The public API may change before 1.0; every breaking
+> change is listed in the [changelog](https://github.com/Pedro-g-Lima/COSMIC_WARP/blob/main/CHANGELOG.md).
+
+## Highlights
+
+- **GPU-parallel** — physics, integration and termination fused into one
+  kernel per step for every environment; `simulate()` replays the step as a
+  CUDA graph
+- **Float64 throughout** — all dynamic state and physics in double precision
+- **Integrators** — Euler, RK4 (default) and a symplectic kick–drift–kick scheme, each
+  verified to reach its nominal order of convergence
+- **Physics** — point-mass gravity, J2, atmospheric drag, gravity-gradient
+  and magnetic torques, each switchable; variable mass from thruster
+  propellant use
+- **Actuators** — up to 8 thrusters (Isp, propellant use) and 4 reaction
+  wheels (momentum saturation, viscous friction) per body
+- **Multi-body** — several spacecraft per environment, configured per body
+  and terminated together (no forces between them)
+- **Differentiable** — reverse-mode gradients through the dynamics w.r.t.
+  initial state, control sequences and vehicle parameters, checked against
+  finite differences
+- **Reinforcement learning** — on-device vectorized environment with
+  auto-reset and zero-copy torch views, adapters for rsl_rl, skrl and
+  rl_games, and a differentiable environment for first-order methods; a
+  reference task and a minimal SHAC trainer in `examples/`
+- **Domain randomization and reset sampling** — on-GPU, including orbital
+  element sampling
+- **Validated** — tests against analytic results: Kepler orbits, J2 secular
+  rates, drag decay, torque-free rigid-body motion, Hohmann transfers, the
+  rocket equation, Clohessy–Wiltshire relative motion, closed-form torques
+
+## Inside a step
+
+One fused kernel per step integrates every entity: one GPU thread each, with
+the forces, torques, actuators, integrator update and termination check
+inside. `simulate()` records the step's kernels once as a CUDA graph and
+replays them.
+
+<picture>
+  <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/Pedro-g-Lima/COSMIC_WARP/main/docs/assets/step-dark.svg">
+  <img alt="One simulate() step: the step kernel, the multi-body reduction, the optional trajectory store and auto-reset, and the trajectory index. Inside the step kernel, per entity: episode counter, terminated check, thruster wrench, integration stages (forces, torques, wheel momenta), store, termination check." src="https://raw.githubusercontent.com/Pedro-g-Lima/COSMIC_WARP/main/docs/assets/step-light.svg">
+</picture>
+
+## Learning through the dynamics
+
+`DiffVecEnv` runs short windows of steps on torch's autograd graph, so a
+loss built from the rewards differentiates through the policy and the
+dynamics at every step — the gradient a first-order method such as SHAC
+uses (example 17).
+
+<picture>
+  <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/Pedro-g-Lima/COSMIC_WARP/main/docs/assets/learning-dark.svg">
+  <img alt="One DiffVecEnv window: H steps of one environment from the carried, detached state; a policy shared by every step maps observations to actions; rewards go into the trainer's loss, and backward runs through every step back to the start of the window." src="https://raw.githubusercontent.com/Pedro-g-Lima/COSMIC_WARP/main/docs/assets/learning-light.svg">
+</picture>
+
+## Installation
+
+```bash
+pip install cosmic-warp
+```
+
+Optional extras (quote the brackets):
+
+| Extra | Adds |
+| --- | --- |
+| `torch` | PyTorch: torch views, differentiable RL |
+| `rsl-rl`, `skrl`, `rl-games` | the trainer adapters (each pulls in torch) |
+| `examples` | matplotlib for the example scripts |
+
+```bash
+pip install "cosmic-warp[torch,skrl]"
+```
+
+The examples and tests live in the repository; to run them, install from a
+clone:
+
+```bash
+git clone https://github.com/Pedro-g-Lima/COSMIC_WARP.git
+cd COSMIC_WARP
+pip install -U pip    # the dependency groups below need pip 25.1
+pip install -e ".[examples]"
+```
+
+A CUDA GPU is the intended target; everything also runs on the CPU
+(`device="cpu"`; the default is `"cuda"`, with no automatic fallback). See
+[Getting started](https://pedro-g-lima.github.io/COSMIC_WARP/getting-started/).
+
+## Quick start
+
+```python
+from cosmic_warp.core.pipeline import SpacecraftSimulator
+
+# 1000 spacecraft in low Earth orbit, drawn from the default reset bounds
+sim = SpacecraftSimulator(num_envs=1000, integrator="rk4", dt=1.0)
+sim.reset()
+
+trajectory = sim.simulate(num_steps=600, save_states=True)
+positions = trajectory["positions"].numpy()   # [num_steps, num_envs, 3]
+```
+
+Whole scenarios (physics, per-body vehicles and actuators, reset bounds,
+initial conditions) can be defined in YAML (this file ships with the
+repository; run from its root):
+
+```python
+from cosmic_warp.core.config import create_simulator_from_yaml
+
+sim = create_simulator_from_yaml("examples/multibody_orbit.yaml")
+```
+
+Continue with [Getting started](https://pedro-g-lima.github.io/COSMIC_WARP/getting-started/)
+and the [YAML reference](https://pedro-g-lima.github.io/COSMIC_WARP/guides/configuration/).
+
+## Examples
+
+Seventeen scripts in [`examples/`](https://github.com/Pedro-g-Lima/COSMIC_WARP/tree/main/examples/), run from the repository root
+(`python examples/<name>.py --help`):
+
+| Topic | Examples |
+| --- | --- |
+| Orbits and integrators | 01 basic orbit, 02 integrator comparison, 05 energy conservation, 08 J2 |
+| Batches | 03 Monte Carlo, 04 scalability, 09 multi-body, 10 domain randomization |
+| Actuators | 06 thrusters, 07 reaction wheels |
+| Gradients | 11 trajectory optimization, 12 system identification |
+| Reinforcement learning | 13 environment, 14–16 PPO with rsl_rl / skrl / rl_games, 17 SHAC |
+
+The [examples gallery](https://pedro-g-lima.github.io/COSMIC_WARP/examples/) shows each
+script with its plots and explanation.
+
+## Documentation
+
+| Section | Contents |
+| --- | --- |
+| [Getting started](https://pedro-g-lima.github.io/COSMIC_WARP/getting-started/) | Install, verify Warp, first orbit |
+| [Guides](https://pedro-g-lima.github.io/COSMIC_WARP/guides/) | Configuration, vehicles, actuators, resets, physics, running, scaling, RL, differentiable simulation |
+| [Examples](https://pedro-g-lima.github.io/COSMIC_WARP/examples/) | One page per example script |
+| [Theory](https://pedro-g-lima.github.io/COSMIC_WARP/theory/) | Conventions, dynamics, gravity and J2, environment, actuators, integrators, sampling, GPU execution, gradients |
+| [Validation](https://pedro-g-lima.github.io/COSMIC_WARP/validation/) | What the test suite checks, against which results |
+| [Benchmarks](https://pedro-g-lima.github.io/COSMIC_WARP/benchmarks/) | Measured throughput per GPU |
+| [API reference](https://pedro-g-lima.github.io/COSMIC_WARP/reference/api/cosmic_warp/) | Generated from the docstrings |
+| [Roadmap](https://pedro-g-lima.github.io/COSMIC_WARP/roadmap/) | Planned for 1.x and beyond |
+
+## Testing and benchmarks
+
+```bash
+pip install -e . --group dev   # pinned test tools (pip >= 25.1)
+pytest                    # full suite on CUDA
+pytest --device cpu       # the same suite without a GPU
+pytest -m "not slow"      # skip the end-to-end training tests
+
+python benchmarks/bench.py --save   # record this GPU's baseline (first run)
+python benchmarks/bench.py          # compare; exit 1 on a >15 % regression
+```
+
+## Contributing
+
+Contributions are welcome — see [CONTRIBUTING.md](https://github.com/Pedro-g-Lima/COSMIC_WARP/blob/main/CONTRIBUTING.md). For
+questions, bugs and feature requests, open an
+[issue](https://github.com/Pedro-g-Lima/COSMIC_WARP/issues).
+
+## Citation
+
+If you use COSMIC WARP in your research, please cite it (GitHub's "Cite this
+repository" uses [`CITATION.cff`](https://github.com/Pedro-g-Lima/COSMIC_WARP/blob/main/CITATION.cff)):
+
+```bibtex
+@software{cosmic_warp,
+  title   = {COSMIC WARP: GPU-Accelerated Spacecraft Simulator},
+  author  = {Lima, Pedro and Arora, Aman},
+  year    = {2026},
+  version = {0.1.0},
+  url     = {https://github.com/Pedro-g-Lima/COSMIC_WARP}
+}
+```
+
+Please also cite the tools it builds on, in particular NVIDIA Warp. BibTeX
+for these and for every source behind the physics models is on the
+[references page](https://pedro-g-lima.github.io/COSMIC_WARP/references/).
+
+## License
+
+[Apache License 2.0](https://github.com/Pedro-g-Lima/COSMIC_WARP/blob/main/LICENSE).
+
+## Acknowledgements
+
+COSMIC WARP is developed by the SpaceR research group at SnT, University of
+Luxembourg.