loco-3d/crocoddyl

Health v2 · maintenance only

Adoption not part of the score

1285 stars · 213 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Crocoddyl is an open-source optimal control library for robot control under contact sequences, computing optimal trajectories and feedback gains using efficient Differential Dynamic Programming (DDP)-like solvers. It is written in C++ with Python bindings and leverages Pinocchio for fast rigid-body dynamics and analytical derivatives.

Use cases

• compute optimal trajectories for legged robots with contact sequences
• implement model predictive control for robotic manipulators
• solve trajectory optimization problems with DDP algorithms
• generate feedback gains for whole-body robot control
• plan dynamic motions like jumping or running for quadrupeds
• prototype optimal control solvers in Python with C++ performance

When to choose

• you need real-time optimal control or MPC for robots with contact dynamics
• you want efficient DDP-based trajectory optimization with analytical derivatives
• you work with Pinocchio and need a compatible optimal control layer
• you need both C++ performance and Python bindings for robotics research

When to avoid

• you need a general-purpose nonlinear optimizer unrelated to robot dynamics
• you want a plug-and-play motion planner without formulating optimal control problems
• your project requires a permissively integrated GUI or simulation environment out of the box

Facets

library · maturity active

simulation robotics math robotics windows python cpp optimal-control differential-dynamic-programming trajectory-optimization model-predictive-control legged-robotics motion-planning pinocchio optimization algorithms research linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="loco-3d/crocoddyl")