Pinocchio is a modern, high-performance library of rigid-body dynamics algorithms, inverse kinematics, and Jacobians for robotics, developed since 2015 at LAAS-CNRS in Toulouse and Inria, by the team of Justin Carpentier, Nicolas Mansard, Olivier Stasse, Florent Lamiraux, and others. BSD-2-Clause license.
Pinocchio is the heart of the Stack of Tasks ecosystem and the reference library for the French school of humanoid robotics (with the HRP, TALOS, and Pyrene robots). It implements Roy Featherstone's algorithms (RNEA — Recursive Newton-Euler Algorithm, CRBA — Composite Rigid Body Algorithm, ABA — Articulated Body Algorithm) with carefully cache-optimized memory layouts. It supports spatial algebra in the form of 6D twists/wrenches.
Key applications: model predictive control (MPC) for humanoids (Crocoddyl, used by Boston Dynamics — unofficially — and ANYbotics), task-space inverse dynamics (TSID), analytical automatic differentiation (CasADi integration) for trajectory optimization, RL with symbolic models. APIs: C++17 (native) and Python (pinocchio package). The standard in European research projects (EUROBENCH, MEMMO, AGIMUS).
An API Library is a software package that exposes programmatic interfaces for communicating with a device, service, or system. In robotics it typically forms a lightweight integration layer built on top of the manufacturer's official API or an open-source project, abstracting low-level protocol details and providing language-native bindings (Python, C++, Java, etc.).
A family of open-source libraries for motion planning (RRT, PRM, Lazy-PRM), inverse kinematics, rigid-body dynamics, and trajectory optimization: MoveIt, OMPL, Pinocchio, Drake, RBDL, TrajOpt, KDL, Bullet, iDynTree.
HRP-4, TALOS, Pyrene, Romeo (PAL Robotics) robots, Solo and Bolt (Open Dynamic Robot Initiative), Cassie and Digit (Agility Robotics — via Crocoddyl), ANYmal (ANYbotics). Built into MPC stacks used on Toyota T-HR3, Honda ASIMO follow-ups, and in many NeurIPS / ICRA / RSS papers.
github.com/stack-of-tasks/pinocchio ~2.4k★, > 100 commits/month, > 60 contributors, > 400 publications citing Pinocchio (Carpentier et al., SII 2019). Dedicated Stack of Tasks Discord ~1.5k.
Boston Dynamics bipedal humanoid robot. The fully electric generation unveiled in 2024 succeeds the hydraulic Atlas that was retired after more than a decade of research.
Tutor Intelligence industrial robot designed for palletizing, depalletizing and case-picking in warehouses and factories, offered as Robotics-as-a-Service from $14/hr.
Unitree H1 is a full-size general-purpose humanoid robot (~180 cm, ~47 kg). Bipedal, 5 DOF per leg + 4 DOF per arm, 3.3 m/s walking speed, 360° perception via 3D LiDAR + depth camera, Unitree M107 PMSM joint motors with ~360 N·m peak knee torque. Standard compute: Intel Core i5/i7; optional NVIDIA Jetson Orin NX.
Bipedal humanoid robot by Unitree Robotics, designed as a compact research, development, and developer platform.
Figure 03 is the third-generation humanoid robot from Figure AI, designed for Helix, home environments, and scalable mass production.
Compact, high-dynamics bipedal humanoid by MagicLab. 140 cm, 40 kg, 24–50 DOF, walking speed up to 2.5 m/s. Unveiled on 8 July 2025 with martial arts and acrobatic demos.
Requires Eigen3, Boost. Python binding via Boost.Python or Eigenpy.
License family: Permissive
Aligator stack, RL-ready bindings, Python JIT compilation.
C++17, new dynamics API, experimental GPU backend.
Contact refactoring, improved MJCF support.
CasADi integration, automatic differentiation, FCL collision.
Full transition to C++14, new spatial algebra API, Python 3 bindings.
First public open-source release from LAAS-CNRS.
