Documentation [ Getting Started | Tutorials | C++ API | Python API | Release Notes ]
NVIDIA cuMotion is a high-performance motion generation library for robotics, focused mainly on manipulation. GPU acceleration is leveraged throughout where beneficial for performance.
Capabilities include:
- Kinematics and inverse kinematics (IK)
- Collision-aware IK
- Collision-aware graph-based path planning
- Collision-aware trajectory optimization and end-to-end motion generation
- Reactive control via the mathematical framework of RMPflow
- Time-optimal (collision-unaware) trajectory generation for paths given by any number of moves in configuration space (joint space) and/or task space
- Collision sphere generation, for approximating a closed mesh by a set of spheres
- Robot segmentation, for removing the contribution of a robot arm from one or more depth image streams
Robots with any number of degrees of freedom are supported.
cuMotion is implemented in C++ and provides a complete set of Python bindings. Although developed as a standalone library, cuMotion will also soon serve as the motion generation backend for Isaac ROS cuMotion and the manipulation reference workflows in Isaac ROS.
cuMotion is descended from two libraries:
- Lula
was released for a number of years as part of Isaac Sim. It has been entirely subsumed by cuMotion,
and Python users of Lula will find that
import cumotion as lulawill suffice for running most client code with only a few modifications. - cuRobo was developed by NVIDIA Research and continues to be maintained as a library for robotics researchers. cuMotion incorporates optimized and hardened implementations of algorithms first introduced in cuRobo, including those for collision-aware IK and trajectory optimization.
See the documentation for a full overview of features.
The current release of the cuMotion library is provided as a set of shared object files and Python wheel files for the following platform configurations.
| Platform | Python | CUDA | Tested OS | Download |
|---|---|---|---|---|
| Jetson Orin | 3.10 | 12.6 | JetPack 6.2.1 | Latest Release |
| Jetson Thor | 3.12 | 13.0 | JetPack 7.0 | Latest Release |
| x86_64 + NVIDIA GPU (Turing or later) | 3.10 | 12.6 | Ubuntu 22.04 | Latest Release |
| x86_64 + NVIDIA GPU (Turing or later) | 3.12 | 13.0 | Ubuntu 24.04 | Latest Release |
Only Linux-based operating systems are supported in the current release.
The software has been tested on Jetson AGX Orin and Jetson AGX Thor but is expected to work on other Jetson Orin configurations (e.g., Jetson Orin NX).
On x86-based computers, an NVIDIA GPU of the Turing generation (e.g., GeForce RTX 2080) or later is required. For optimal performance on Blackwell GPUs (e.g., RTX PRO 6000, GeForce RTX 5080), the CUDA 13 variant of the cuMotion package is recommended. See the CUDA Toolkit Release Notes for details on the minimum GPU driver version required for a given CUDA version.
Download the cuMotion package for your desired platform from Releases and extract the contents:
tar xzvf cumotion-<version>-<python_version>-<cuda_version>-<arch>.tar.gz && \
cd cumotion-<version>-<python_version>-<cuda_version>-<arch>
See Getting Started for more details.
Copyright © 2019-2025 NVIDIA Corporation & Affiliates