coal-library
An extension of the Flexible Collision Library
Decision gist · record as of 2026-08-14
Yes, if you need collision detection and distance computation for robotics or 3D geometric applications. Coal offers state-of-the-art GJK/EPA performance, is actively maintained, has no known vulnerabilities, and integrates well with established robotics frameworks. The medium install friction (compiled dependencies) is manageable via conda. Not necessary if you only need basic bounding-box overlap checks or are already embedded in a framework that provides collision detection.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires compiled dependencies (cmeel-boost, cmeel-assimp, cmeel-octomap, cmeel-qhull, eigenpy); conda installation recommended to handle binary wheels automatically.
- Medium install friction: depends on six compiled packages (cmeel, cmeel-assimp, cmeel-boost, cmeel-octomap, cmeel-qhull, eigenpy).
- Pre-built wheels available for multiple Python versions on macOS and Linux.
License · maintenance · safety
BSD-3-Clause (permissive) — BSD-3-Clause permissive license allows commercial and private use with minimal restrictions; attribution and license notice required in distributions.
last release 2025-02-12 (548 days) · last repo commit 2026-05-21 · 1 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 180,243 downloads/mo, #10,148 on PyPI
Alternatives
Verify before relying
import coal
# Create shapes
shape1 = coal.Ellipsoid(0.7, 1.0, 0.8)
shape2 = coal.Sphere(0.5)
# Define placements
T1 = coal.Transform3s()
T2 = coal.Transform3s()
# Collision request and result
request = coal.CollisionRequest()
result = coal.CollisionResult()
# Perform collision check
coal.collide(shape1, T1, shape2, T2, request, result)- Whether Python bindings cover the full C++ API surface or a subset of collision/distance operations.
- Whether contact patch computation is exposed in the Python API or only in C++.
- Performance characteristics of accelerated GJK variants relative to baseline GJK in typical use cases.
What it is and what it does
Coal is a collision detection and distance computation library for 3D rigid bodies, forked and substantially rewritten from the Flexible Collision Library (FCL) since 2015 and renamed in 2024. It implements optimized GJK and EPA algorithms for narrow-phase collision detection, supports safety margins, computes contact points and patches, and handles a wide range of geometries including primitives (boxes, spheres, capsules, ellipsoids, cones), convex meshes, bounding volume hierarchies, height fields, and octrees.
The library is used in robotics frameworks like Pinocchio, the Humanoid Path Planner, and the Simple simulator. It provides Python bindings for prototyping and integration, making it accessible to researchers and engineers who need fast, reliable collision queries without implementing low-level geometric algorithms. The package requires six compiled dependencies (cmeel, cmeel-assimp, cmeel-boost, cmeel-octomap, cmeel-qhull, eigenpy) and is actively maintained with wheels for modern Python versions on macOS and Linux.
Use it for
- Motion planning for humanoid robots: check collisions between planned trajectories and obstacles or the robot itself.
- Physics simulation: detect contacts and compute contact patches for constraint-based rigid body dynamics.
- Grasp planning: compute distances and contact points between gripper and object geometries.
- Path validation: validate collision-free paths in high-dimensional configuration spaces during planning.
- Proximity queries: compute lower bounds on distances between objects for safety margins in real-time control.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you need collision detection and distance computation for robotics or 3D geometric applications.
Coal offers state-of-the-art GJK/EPA performance, is actively maintained, has no known vulnerabilities, and integrates well with established robotics frameworks. The medium install friction (compiled dependencies) is manageable via conda. Not necessary if you only need basic bounding-box overlap checks or are already embedded in a framework that provides collision detection.
Install
coal-library on PyPI
Before you install
Medium install friction: depends on six compiled packages (cmeel, cmeel-assimp, cmeel-boost, cmeel-octomap, cmeel-qhull, eigenpy). Pre-built wheels available for multiple Python versions on macOS and Linux. Actively maintained with recent commits.
Requires compiled dependencies (cmeel-boost, cmeel-assimp, cmeel-octomap, cmeel-qhull, eigenpy); conda installation recommended to handle binary wheels automatically.
License in practice
BSD-3-Clause permissive license allows commercial and private use with minimal restrictions; attribution and license notice required in distributions.
Quickstart
import coal
# Create shapes
shape1 = coal.Ellipsoid(0.7, 1.0, 0.8)
shape2 = coal.Sphere(0.5)
# Define placements
T1 = coal.Transform3s()
T2 = coal.Transform3s()
# Collision request and result
request = coal.CollisionRequest()
result = coal.CollisionResult()
# Perform collision check
coal.collide(shape1, T1, shape2, T2, request, result)Verify before relying
- Whether Python bindings cover the full C++ API surface or a subset of collision/distance operations.
- Whether contact patch computation is exposed in the Python API or only in C++.
- Performance characteristics of accelerated GJK variants relative to baseline GJK in typical use cases.
Package facts
| License | BSD-3-Clause permissive |
| Python support | Supports the current Python release >=3.8 |
| Install friction | Medium. Platform-specific wheel |
| Runtime dependencies | 6 packagescmeelcmeel-assimpcmeel-boostcmeel-octomapcmeel-qhulleigenpy |
| Maintenance | Actively maintained 548 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 180,243 / month, #10,148 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
Evidence: coal_library-3.0.1-0-cp310-cp310-macosx_10_9_x86_64.whl; coal_library-3.0.1-0-cp310-cp310-macosx_11_0_arm64.whl; coal_library-3.0.1-0-cp310-cp310-manylinux_2_28_aarch64.whl; coal_library-3.0.1-0-cp310-cp310-manylinux_2_28_x86_64.whl; coal_library-3.0.1-0-cp311-cp311-macosx_10_9_x86_64.whl; coal_library-3.0.1-0-cp311-cp311-macosx_11_0_arm64.whl; coal_library-3.0.1-0-cp311-cp311-manylinux_2_28_aarch64.whl; coal_library-3.0.1-0-cp311-cp311-manylinux_2_28_x86_64.whl; coal_library-3.0.1-0-cp312-cp312-macosx_10_9_x86_64.whl; coal_library-3.0.1-0-cp312-cp312-macosx_11_0_arm64.whl; coal_library-3.0.1-0-cp312-cp312-manylinux_2_28_aarch64.whl; coal_library-3.0.1-0-cp312-cp312-manylinux_2_28_x86_64.whl; coal_library-3.0.1-0-cp313-cp313-macosx_10_9_x86_64.whl; coal_library-3.0.1-0-cp313-cp313-macosx_11_0_arm64.whl; coal_library-3.0.1-0-cp313-cp313-manylinux_2_28_aarch64.whl; coal_library-3.0.1-0-cp313-cp313-manylinux_2_28_x86_64.whl; coal_library-3.0.1-0-cp38-cp38-macosx_10_9_x86_64.whl; coal_library-3.0.1-0-cp38-cp38-manylinux_2_28_aarch64.whl; coal_library-3.0.1-0-cp38-cp38-manylinux_2_28_x86_64.whl; coal_library-3.0.1-0-cp39-cp39-macosx_10_9_x86_64.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “distance computation shapes”
- coal-libraryCoal provides collision detection, distance computation, and contact…
- python-fclPython bindings for the Flexible Collision Library (FCL) that perform…
- python-LevenshteinComputes Levenshtein edit distance, string similarity, and…
Give your agent the search over MCP, or paste the wish link into any chat.
More Scientific/Engineering packages
NumPy provides an N-dimensional array object and a comprehensive suite of mathematical, linear algebra, Fourier transform, and random number functions for scientific computing in Python.
pandas provides fast, flexible data structures (Series and DataFrame) for loading, cleaning, transforming, and analyzing labeled or relational data in Python.
scipy provides numerical algorithms for mathematics, science, and engineering—including optimization, integration, linear algebra, Fourier transforms, signal and image processing, and ODE solvers—built on numpy arrays.
scikit-learn provides a comprehensive Python library for supervised and unsupervised machine learning, including classification, regression, clustering, dimensionality reduction, and model evaluation tools built on NumPy and SciPy.
Install it if you need to train, evaluate, or deploy supervised or unsupervised learning models.
dill extends Python's pickle module to serialize and deserialize a much wider range of Python objects, including functions, lambdas, classes, and interpreter sessions, to byte streams for storage or network transmission.
Multiprocess is an enhanced fork of Python's standard multiprocessing library that uses dill for better serialization, allowing you to spawn processes with a threading-like API and share complex objects between them.
Install it if you use multiprocessing and encounter pickle serialization limits with lambdas or complex objects.
See also coal · ifcopenshell · libcoal · python-fcl · libpinocchio · pin · mplib · rsl-rl-lib · cmeel-qhull · trimesh