ompl/ompl

The Open Motion Planning Library (OMPL)

What it solves

OMPL provides a standardized library for motion planning, allowing robots to find valid paths from a start point to a goal while avoiding obstacles. It eliminates the need for developers to implement complex sampling-based planning algorithms from scratch.

How it works

The library implements over 40 sampling-based planning algorithms (such as RRT-Connect, PRM, and RRT*) across more than 20 different state spaces, including Euclidean space and SE(3). It supports both C++ and Python and can be accelerated using VAMP (Vector-Accelerated Motion Planning) for SIMD-optimized collision checking to achieve millisecond planning speeds.

Who it’s for

It is designed for robotics engineers and researchers who need to implement motion planning in C++ or Python and require a wide variety of pre-implemented algorithms and state spaces.

Highlights

  • Over 40 sampling-based planning algorithms implemented.
  • Support for 20+ state spaces.
  • SIMD-accelerated planning via VAMP for high performance.
  • Extensible to custom planners and state spaces in C++ and Python.

Related

  • Project
    KavrakiLab/vamp

    A hardware-accelerated motion planning library that uses CPU SIMD instructions to perform collision checking and forward kinematics in microseconds.

  • Project
    hanruihua/RDA-planner

    An optimization-based MPC motion planner that uses ADMM to enable real-time, collision-free navigation for robots in cluttered environments.

  • Project
    open-planning/roboplan

    A modern C++ robot motion planning library based on Pinocchio that provides tools for inverse kinematics, RRT-based path planning, and trajectory timing.

  • Project
    uxlfoundation/oneDAL

    A C++ and DPC++ library that provides hardware-accelerated machine learning routines for tabular data across CPUs, GPUs, and distributed systems.

  • Project
    microsoft/FLAML

    A lightweight Python library for efficient automated machine learning (AutoML) and hyperparameter tuning to find high-quality models with minimal computational resources.