Draft:Open-Source Leg

  1. Home
  2. Wiki
  • Comment: This shows the hallmarks of artificial intelligence writing. I concur with others that this shows clear signs of AI writing and should not be accepted. Including the rule of three, superficial notability analysis, excessive lists, and hosts of other small signs, especially structural elements of the article. aaronneallucas (talk) 04:27, 6 December 2025 (UTC)
  • Comment: This still feels like it might be AI. Lots of suspicious sentences like "A public forum facilitates discussion and troubleshooting" and "The project has received coverage from international media, including Economy Chosun, which highlighted...", and too many bulleted lists. —pythoncoder (talk | contribs) 02:54, 6 December 2025 (UTC)
Open-Source Leg
DeveloperElliott J. Rouse
Initial releaseOctober 2020 (2020-10)
Stable release
2.5 / July 2024 (2024-07)
Repositoryhttps://github.com/neurobionics/opensourceleg
Written inPython, C
Operating systemLinux, ROS 2
LicenseGPLv3 (software)
Apache 2.0 (hardware)
Websitehttps://opensourceleg.org

The Open-Source Leg (OSL) is an open-source robotic knee–ankle prosthesis designed for research in powered prosthetic control, gait biomechanics, and wearable robotics. The platform provides openly licensed mechanical designs, electronics schematics, firmware, and software libraries intended to support reproducible experiments and cross-laboratory comparison.[1]

The first scientific description of the system appeared in 2020 in Nature Biomedical Engineering, which detailed the OSL’s mechanical design, sensing systems, control structure, and initial clinical evaluation.[1] The project has received coverage from international media, including Economy Chosun, which highlighted the platform’s potential to standardize prosthetic research and expand access to experimental bionic technology.[2] Development has been supported by multiple awards from the U.S. National Science Foundation.[3][4][5][6]

History

[edit]

The OSL originated under NSF National Robotics Initiative award #1734586 (2017–2020), which supported foundational development of the mechanical hardware, embedded control electronics, sensing integration, and open-source dissemination.[3] A University of Michigan news release introduced the system publicly in 2019.[7]

The platform’s first peer-reviewed scientific description was published in 2020 in Nature Biomedical Engineering, presenting detailed schematics, actuator characterization, sensing architecture, and clinical testing results.[1]

Between 2020 and 2022, the NSF NRI:INT collaborative research award #2024237 supported development of continuous-torque control methods, benchmarking protocols, and multi-laboratory controller evaluation.[4]

In 2024, the University of Michigan Robotics Department reported on ecosystem-building efforts for the OSL, including open-source governance, documentation development, community infrastructure, and partnerships with external laboratories.[8]

The project continues under the NSF Pathways to Open-Source Ecosystems (POSE) program:

  • POSE Phase I – #2229418 (2022–2023) – governance, documentation systems, and contributor pathways.[5]
  • POSE Phase II – #2315895 (2023–2026) – sustainability planning, onboarding tools, safety documentation, and long-term ecosystem maintenance.[6]

The platform has been adopted by research groups in North America and Europe, including clinical partners such as the Shirley Ryan AbilityLab (SRALab).[9]

Design

[edit]

Hardware

[edit]

The OSL consists of modular powered knee and ankle joints that share a similar mechanical structure, simplifying assembly and repair across research laboratories.[1] Both joints use high-torque brushless DC motors originally developed for aerial robotics, chosen for torque density, low rotor inertia, and suitability for backdrivable actuation.

The joints employ low-ratio belt transmissions to increase backdrivability and reduce passive impedance. The ankle uses a two-stage reduction, while the knee uses a configurable single- or dual-stage design.[1]

A selectable series elastic element can be included in the drivetrain, providing tunable joint stiffness for experimental investigation. Best et al. (2024) characterized torsion-based elastic actuation strategies compatible with the OSL’s modular architecture.[10]

Integrated sensing includes:

  • magnetic encoders for joint and motor position
  • multi-axis load cells for torque or ground-reaction force estimation
  • inertial measurement units (IMUs) for segment kinematics
  • temperature, voltage, and current sensors for actuator protection

The electronics platform includes a six-channel data acquisition system, high-frequency analog sampling, and digital communication for real-time motor control. Shetty et al. (2022) used the sensing and electronics architecture to conduct actuator system identification and evaluation of torque dynamics.[11]

The assembled system weighs under 6 kg and uses machined aluminum components with standardized mounting points. It supports tethered bench testing and untethered battery-powered locomotion.[1]

Software

[edit]

The OSL software includes embedded firmware, mid-level joint controllers, and high-level Python interfaces for experiment scripting.[12]

Embedded firmware manages:

  • motor voltage and current control
  • encoder and load-cell sampling
  • sensor fusion and filtering
  • motor-driver communication
  • safety monitoring and watchdog functions

Mid-level controllers implement:

  • voltage control
  • current (torque) control
  • position control
  • impedance control for biological stiffness and damping emulation

The continuous-torque controller framework developed under NSF award #2024237 supports smooth transitions between gait phases and works with finite-state machine controllers and adaptive impedance strategies.[4]

The Python API offers:

  • gait event detection
  • parameter tuning
  • real-time data logging
  • visualization and debugging utilities
  • ROS 2 integration

Research by Harris et al. (2024) and Bolívar-Nieto et al. (2021) evaluated prosthesis control methods—such as torque-based and impedance-based strategies—that are compatible with the OSL’s control and sensing architecture.[13][14]

Software releases are distributed through GitHub and PyPI with automated testing for reproducibility.[12]

Research use

[edit]

The OSL is cited in research involving prosthesis control evaluation, gait biomechanics, and robotic actuation. Publications using methodologies compatible with the OSL architecture include:

  • Best et al. (2024) – evaluation of torsion-based elastic actuation methods relevant to modular prosthesis actuation.[10]
  • Harris et al. (2024) – assessment of knee–ankle control strategies, including impedance-based approaches.[13]
  • Bolívar-Nieto et al. (2021) – modeling and control of impedance-based prosthesis controllers.[14]
  • Shetty et al. (2022) – actuator system identification using sensing and electronics architectures compatible with OSL hardware.[11]

Community and adoption

[edit]

The OSL ecosystem includes CAD models, electronics schematics, firmware, control libraries, documentation, and community-support resources. A public forum facilitates discussion and troubleshooting. Research groups internationally use the platform for gait biomechanics, prosthetic control, and wearable robotics research.

In 2021, Humotech partnered with the project to offer assembled OSL units for laboratories without in-house fabrication capabilities.[15]

See also

[edit]

References

[edit]
  1. ^ a b c d e f Azocar, A. F.; Rouse, E. J. (2020). "Design and clinical implementation of an open-source bionic leg". Nature Biomedical Engineering. 4 (10): 941–953. doi:10.1038/s41551-020-00619-3. PMC 7581510. PMID 33020601.
  2. ^ "The Open-Source Leg Project". Economy Chosun. 2025-04-25.
  3. ^ a b "NSF Award #1734586 – An Open-Source Robotic Leg Platform that Lowers the Barrier for Advanced Prosthetics Research". NSF.gov.
  4. ^ a b c "NRI: INT Collaborative Research – Open-Source Framework for Continuous Torque Control of Prosthetic Legs". IRAD.
  5. ^ a b "POSE Phase I – Advancement of an Open-Source Hardware and Software Ecosystem". Elsevier Pure.
  6. ^ a b "POSE Phase II – Continued Progression of an Open-Source Hardware & Software Ecosystem". Elsevier Pure.
  7. ^ "Open-source bionic leg aims to advance prosthetics research". University of Michigan News. 2019-06-04.
  8. ^ "Building an Ecosystem for the Open-Source Leg". University of Michigan Robotics. 2024.
  9. ^ "Open-Source Bionic Leg Project". Shirley Ryan AbilityLab. 20 August 2019.
  10. ^ a b Best, T. K. (2024). "A Compact, Two-Part Torsion Spring Architecture". IEEE/ASME Transactions on Mechatronics. doi:10.1109/TMECH.2024.3508469.
  11. ^ a b Shetty, V. S. (2022). "System identification for wearable robotic actuators". IEEE Robotics and Automation Letters. doi:10.1109/LRA.2022.3144790.
  12. ^ a b "Open-Source Leg Software Repository". GitHub.
  13. ^ a b Harris, I. R. (2024). "Evaluation of powered knee–ankle prosthesis control". IEEE Robotics and Automation Letters. doi:10.1109/LRA.2024.3416769.
  14. ^ a b Bolívar-Nieto, E. A. (2021). "Powered prosthesis control modeling". Mechatronics. doi:10.1016/j.mechatronics.2021.102635.
  15. ^ "U-M, Humotech partner to bring open-source bionic leg to research labs". Humotech. 2021-12-16.