[ompl-users] [software] New release (1.7.0) of the Open Motion Planning Library (OMPL)

[Lydia K. <ka...@ri...>]

Hi all,

We are happy to announce version 1.7.0 of the Open Motion Planning Library (OMPL) <http://ompl.kavrakilab.org/>.

This release introduces several significant enhancements and changes aimed at improving functionality, compatibility, and user experience. Below is a summary of the main updates:​

New Planners and State Spaces:

Effort Informed Trees (EIT*): A new planner designed to efficiently handle kinodynamic planning problems by leveraging an informed search strategy.​
Dubins State Spaces: Three new 3D extensions to the Dubins model have been added: Vana, Owens, and Vana-Owens state spaces.
RRT-Rope: The rope short-cutting technique from RRT-Rope has been added to our PathSimplifier class.

Removed Features:

ODE and MORSE Extensions: Support for the ODE (Open Dynamics Engine) and MORSE robot simulator has been removed.

Performance Improvements:

Dubins State Space Optimization: Implemented Dubins Set Classification for faster distance computations in the Dubin State Space.

System Requirements:

C++17 Compliance: A C++17 compliant compiler is now required to build OMPL.
CMake and Boost Versions: The minimum required version of CMake is now 3.5, and the minimum supported version of Boost is 1.68.​
NumPy Version: The minimum required version of NumPy for Python bindings is now 2.0.

Build and Deployment:

Docker Support: Build targets have been added for creating Docker images on Ubuntu 24.04 for OMPL, the PlannerArena web server, and the OMPL web app. These Docker images are available on Docker Hub <https://hub.docker.com/u/kavrakilab>.
Python Wheels: Python wheels <https://github.com/ompl/ompl/releases> for OMPL are now available, simplifying the installation process for Python users.

Contributions to this release were made by (in alphabetical order):
Kyle Cesare, independent
Gaël Écorchard, Czech Technical University in Prague
Ryan Friedman, Aerovironment, Inc.
Jonathan Gammell, Queen’s University
Michael Görner, University of Hamburg
Valentin Hartmann, University of Stuttgart
Wolfgang Hönig, Technical University Berlin
Taylan İşleyici, Middle East Technical University
Zak Kingston, Purdue University
Jaeyoung Lim, ETHZ
Rhys Mainwaring, independent
Mark Moll, Metron, Inc.
Johnny Nunez, University of Barcelona
Andreas Orthey, Realtime Robotics
Louis Petit, University of Sherbrooke
Yuri Rocha, MakinaRocks
Marlin Strub, Gravis Robotics
Wil Thomason, RAI Institute
Jafar Uruç, Humanoid

Download
You can download OMPL and find more information about OMPL on its homepage at http://ompl.kavrakilab.org <http://ompl.kavrakilab.org/>. 

What is next?
We have already started work on OMPL 2.0. Some features that will be available in our repository <https://github.com/ompl/ompl> soon include:
Example integrations that show how OMPL can be used in combination with MuJoCo, Bullet, Pinocchio and other robotics frameworks.
New Python bindings based on nanobind <https://github.com/wjakob/nanobind>.
A revamped web page to showcase the broad range of projects where OMPL has been used.
Our next big goal is to integrate the ideas of  VAMP <https://www.kavrakilab.org/publications/thomason2024vamp.pdf>’s SIMD-accelerated planning with OMPL, which should result in orders of magnitude speed-up.

Of course, we welcome contributions from others, too!

With best wishes,

Lydia Kavraki 



Lydia E. Kavraki, Ph.D.
Kenneth and Audrey Kennedy Professor of Computing
Professor of Computer Science
Professor of Bioengineering
Professor of Computer and Electrical Engineering
Professor of Mechanical Engineering
Director, Ken Kennedy Institute
Rice University
ka...@ri... <mailto:ka...@ri...>
(713) 348-5737
https://profiles.rice.edu/faculty/lydia-e-kavraki
http://www.kavrakilab.org <http://www.kavrakilab.org/>
http://kenkennedy.rice.edu <http://kenkennedy.rice.edu/>