Edge Rewrite
location Source code View original
// request.cf · coarse context

A page that knows where it met you.

Only coarse request metadata is shown. This demo does not display or persist visitor IP addresses.

Country
US
Cloudflare location
CMH
Connection
HTTP/2
Language
Not provided

Ray ID: a21b363cac32de81

Jump to content

Draft:MoveIt

From Wikipedia, the free encyclopedia
  • Comment: In accordance with the Wikimedia Foundation's Terms of Use, I disclose that I have been paid by my employer for my contributions to this article. Dave (talk) 19:46, 18 July 2026 (UTC)


MoveIt
Original authorsIoan Șucan, Sachin Chitta
DeveloperMoveIt community
Release2012; 14 years ago (2012)[1]
Written inC++, Python
Operating systemLinux
PlatformRobot Operating System
TypeMotion planning framework
LicenseBSD 3-Clause[2]
Websitemoveit.ai

MoveIt (formerly stylized MoveIt!) is an open-source motion planning framework for robotic manipulators built on the Robot Operating System (ROS). It integrates motion planning, collision detection, kinematics, 3D perception, and trajectory execution into a unified framework, so that a given manipulator is supported through robot-specific configuration rather than application-specific code.[1][3] MoveIt originated at Willow Garage in 2011. Peer-reviewed literature has described it as the primary software library for motion planning and mobile manipulation in ROS,[4] and it is released under the BSD 3-Clause license.[2][4]

History

[edit]

MoveIt originated at Willow Garage, the robotics research lab that also created ROS, as the successor to the lab's earlier "arm navigation" and grasping-pipeline software.[5][6] Ioan Șucan and Sachin Chitta, who had led that predecessor work with E. Gil Jones, began MoveIt itself in October 2011, joined by contributors including Jones, Acorn Pooley, Suat Gedikli, and David Hershberger.[5] The framework was introduced publicly in a 2012 column by Chitta, Șucan, and Steve Cousins in IEEE Robotics & Automation Magazine, published by the IEEE Robotics and Automation Society.[1]

After Willow Garage wound down, SRI International supported MoveIt's development from October 2013 to June 2015.[5] Since 2015 the project has been maintained by PickNik Robotics, a Boulder, Colorado company co-founded by Dave Coleman, the project's lead maintainer and first author of a 2014 study of the framework's adoption, together with a community of external contributors from organizations including Delft University of Technology and Southwest Research Institute.[5][3][7][8]

Design

[edit]

MoveIt is designed to be independent of any particular robot. A robot is described by a URDF model together with configuration generated by a graphical Setup Assistant, after which the framework provides motion planning, kinematics, and collision checking for that robot without robot-specific application code.[3][9] Three concepts are central to its architecture.

The planning scene is MoveIt's representation of the robot and its environment. It combines the robot's kinematic state with a model of the surrounding world built from 3D sensor data — typically an octree-based occupancy map — and known collision objects, and it is kept updated so that plans are validated against the current environment.[1][3]

The planning pipeline processes a motion request through a planning algorithm together with a series of modular processing stages that run before and after it. These stages, termed planning request adapters, perform tasks such as validating the request and applying time parameterization or smoothing to the resulting trajectory before execution.[3]

A plugin architecture lets the major components be swapped without changing application code. Rather than implementing a single planning algorithm, MoveIt exposes a planner plugin interface whose default backend is the Open Motion Planning Library (OMPL) of sampling-based planners; separate plugin interfaces cover inverse kinematics solvers and collision checkers.[3] MoveIt integrates with the RViz visualization tool for interactive planning and with ROS control frameworks for trajectory execution on hardware.[3]

This extensibility was significant to the framework's uptake. Because planners, kinematics solvers, and collision checkers are selected through configuration rather than compiled in, users are not committed to a single motion planning algorithm and researchers can substitute their own implementations while reusing the surrounding infrastructure. The framework's developers identified this plugin design, alongside the automated setup tooling, as a principal reason the software lowered the barrier to entry for robotic manipulation.[3] Later research tooling has built on the same property to develop and evaluate new planners.[10]

MoveIt 2

[edit]

MoveIt 2 is the framework's implementation for ROS 2 and is the project's actively developed branch.[2] A beta was announced in February 2020, intended to enable faster and more reactive motion planning through real-time control.[7] A release for the ROS 2 "Humble" long-term-support distribution followed in 2022, adding features including hybrid planning and jerk-limited trajectory smoothing via the Ruckig library.[11] ROS 1 itself reached end of life with its final distribution, Noetic Ninjemys, on 31 May 2025.[12]

Adoption

[edit]

MoveIt was developed initially on Willow Garage's PR2 robot.[1][3] It has since been used on other platforms, including NASA's Robonaut program,[7] and MoveIt-based tooling has been applied to motion planning problems for Robonaut 2 such as traversing handrails aboard the International Space Station.[10] A 2023 review reported that MoveIt was used on more than 150 robots spanning vendors including NASA, ABB, Yaskawa, and Fanuc, and identified it as the core motion planning library for the ROS-Industrial project.[4]

The ROS Metrics project recorded moveit-kinematics and moveit as the fourth and fifth most-downloaded ROS packages first introduced in 2017.[13] Trade-press coverage of the MoveIt 2 release listed companies using the framework, among them Fetch Robotics, Franka Emika, PAL Robotics, Realtime Robotics, Acutronic Robotics, and Kinova.[7]

NASA has funded projects extending MoveIt-based software for robotic manipulation in space, including through Small Business Innovation Research awards to PickNik,[14] and third parties have extended the framework with additional manipulation functions for industrial applications.[4]

Reception

[edit]

Researchers have characterized MoveIt as a popular library for motion planning and have built further tooling on top of it for developing new planners, evaluating existing ones, and solving complex planning problems.[10] Its integration with OMPL made it a common route by which sampling-based planning algorithms reached applied robotics work.[3] Peer-reviewed extensions include the MoveIt Task Constructor, which adds task-level planning for manipulation actions composed of multiple interdependent stages.[15] A 2023 review in Robotics and Computer-Integrated Manufacturing described MoveIt as the primary software library for motion planning and mobile manipulation in ROS, attributing its uptake to its reliability, its permissive licensing, and an active user community.[4] Reported usage grew over time: the framework's original developer counted more than 65 robots in 2016,[9] and the 2023 review counted more than 150.[4]

The same review identified gaps in the framework's industrial coverage, finding that features required for many manufacturing applications — among them automatic tool changing and direct control of end-effector speed — were not implemented, and that MoveIt's default sampling-based planners could produce trajectories that appeared unintuitive because the end effector's path between poses is unconstrained. The authors developed and released extensions addressing these gaps, noting that the project relies on community contributions for such capabilities.[4]

Community

[edit]

MoveIt is developed openly on GitHub by a community of contributors drawn from industry and academia.[2] In April 2024 the project's repositories were moved from the ros-planning GitHub organization to a dedicated moveit organization managed by PickNik Robotics, and its website moved from moveit.ros.org to moveit.ai; the announcement noted that MoveIt, unlike ROS itself, is not managed by the Open Source Robotics Foundation.[16] The project has hosted World MoveIt Day, a community hackathon, since 2016.[17]

PickNik Robotics markets a commercial platform built on MoveIt, originally called MoveIt Studio and renamed MoveIt Pro in 2024.[18]

See also

[edit]

References

[edit]
  1. ^ a b c d e Chitta, Sachin; Sucan, Ioan A.; Cousins, Steve (2012). "MoveIt! [ROS Topics]". IEEE Robotics & Automation Magazine. 19 (1): 18–19. doi:10.1109/MRA.2011.2181749.
  2. ^ a b c d "moveit/moveit2". GitHub. Retrieved 2026-07-17.
  3. ^ a b c d e f g h i j Coleman, David; Șucan, Ioan A.; Chitta, Sachin; Correll, Nikolaus (2014). "Reducing the Barrier to Entry of Complex Robotic Software: a MoveIt! Case Study". Journal of Software Engineering for Robotics. 5 (1): 3–16. arXiv:1404.3785. doi:10.6092/JOSER_2014_05_01_p3.
  4. ^ a b c d e f g Malvido Fresnillo, Pablo; Vasudevan, Saigopal; Mohammed, Wael M.; Martinez Lastra, Jose L.; Perez Garcia, Jose A. (2023). "Extending the motion planning framework—MoveIt with advanced manipulation functions for industrial applications". Robotics and Computer-Integrated Manufacturing. 83 102559. doi:10.1016/j.rcim.2023.102559.
  5. ^ a b c d "About". MoveIt. Retrieved 2026-07-17.
  6. ^ "Software". Ioan Alexandru Șucan (personal site). Retrieved 2026-07-17.
  7. ^ a b c d Demaitre, Eugene (February 19, 2020). "MoveIt 2 enables realtime robot arm control with ROS 2". The Robot Report. Retrieved 2026-07-17.
  8. ^ Ryan, Kevin J. (2021). "This Startup Is Tackling One of the Biggest Challenges in Robotics". Inc. Retrieved 2026-07-17.
  9. ^ a b Chitta, Sachin (2016). "MoveIt!: An Introduction". In Koubaa, Anis (ed.). Robot Operating System (ROS): The Complete Reference. Studies in Computational Intelligence. Vol. 625. Cham: Springer. pp. 3–27. doi:10.1007/978-3-319-26054-9_1. ISBN 978-3-319-26052-5.
  10. ^ a b c Kingston, Zachary; Kavraki, Lydia E. (2022). "Robowflex: Robot Motion Planning with MoveIt Made Easy". 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). pp. 3108–3114. arXiv:2103.12826. doi:10.1109/IROS47612.2022.9981698.
  11. ^ "MoveIt 2 Humble Release". MoveIt. 2022-06-02. Retrieved 2026-07-17.
  12. ^ "ROS Noetic End-of-Life: May 31, 2025". Open Robotics Discourse. Open Source Robotics Foundation. Retrieved 2026-07-18.
  13. ^ Lu, David V. "Most Downloaded Packages". ROS Metrics. Retrieved 2026-07-17.
  14. ^ "NASA 'Arms' Astronauts, Industry with Robotic Intelligence". NASA Spinoff. Retrieved 2026-07-17.
  15. ^ Görner, Michael; Haschke, Robert; Ritter, Helge; Zhang, Jianwei (2019). "MoveIt! Task Constructor for Task-Level Motion Planning". 2019 International Conference on Robotics and Automation (ICRA). pp. 190–196. doi:10.1109/ICRA.2019.8793898.
  16. ^ "Move of Nav2 and MoveIt repositories at GitHub". Open Robotics Discourse. April 29, 2024. Retrieved 2026-07-18.
  17. ^ "World MoveIt Day". MoveIt. Retrieved 2026-07-17.
  18. ^ "PickNik Robotics' MoveIt Studio is now MoveIt Pro". The Robot Report. February 22, 2024. Retrieved 2026-07-17.
[edit]

Category:Robotics software Category:Free software programmed in C++ Category:Software using the BSD license Category:Robot Operating System Category:2012 software

Retrieved from "https://en.wikipedia.org/w/index.php?title=Draft:MoveIt&oldid=1365055054"