stack-of-tasks/pinocchio

Health v2 · maintenance only

Adoption not part of the score

3682 stars · 570 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Pinocchio is a fast C++ library (with Python bindings) implementing state-of-the-art rigid body dynamics algorithms for poly-articulated systems, including inverse/forward dynamics, Jacobians, and contact resolution. It provides analytical derivatives of core algorithms, enabling gradient-based optimization, control, and learning in robotics, simulation, and biomechanics.

Use cases

• compute inverse and forward dynamics of a robot model
• calculate analytical derivatives of rigid body dynamics for gradient-based control
• parse URDF robot models and compute kinematics and Jacobians
• simulate articulated mechanisms with frictional contacts
• build model predictive control or reinforcement learning pipelines for legged robots
• compute center of mass and momentum dynamics for biomechanics analysis

When to choose

• you need fast, efficient rigid body dynamics computations in C++ or Python
• you require analytical derivatives of dynamics algorithms for optimization or learning
• you are working on robot control, motion planning, or physics-based simulation
• you need closed-loop kinematic mechanisms and contact modeling support

When to avoid

• you need a full physics engine with out-of-the-box rendering and game features
• you only need simple 2D physics simulation without articulated models
• you want a high-level robotics framework with built-in perception and planning stacks

Facets

library · maturity active

simulation math parser robotics simulation computer-vision graphics cpp python windows cross-platform rigid-body-dynamics kinematics inverse-dynamics forward-dynamics analytical-derivatives urdf automatic-differentiation motion-planning robot-control featherstone-algorithms collision-detection biomechanics physics code-generation linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="stack-of-tasks/pinocchio")