References

Academic References

This section lists key academic papers and resources that influenced MJINX’s design and implementation:

Citations

This project builds on the following academic works:

[1]

Aaron D Ames, Samuel Coogan, Magnus Egerstedt, Gennaro Notomista, Koushil Sreenath, and Paulo Tabuada. Control barrier functions: theory and applications. In 2019 18th European control conference (ECC), 3420–3431. Ieee, 2019.

[2]

James Bradbury, Roy Frostig, Peter Hawkins, Matthew J Johnson, Chris Leary, Dougal Maclaurin, and Skye Wanderman-Milne. JAX: composable transformations of Python+NumPy programs. https://github.com/google/jax, 2018.

[3]

Lukas Brunke, Siqi Zhou, Mingxuan Che, and Angela P Schoellig. Practical considerations for discrete-time implementations of continuous-time control barrier function-based safety filters. In 2024 American Control Conference (ACC), 272–278. IEEE, 2024.

[4]

Andrea Del Prete. Joint position and velocity bounds in discrete-time acceleration/torque control of robot manipulators. IEEE Robotics and Automation Letters, 3(1):281–288, 2018.

[5]

Brian E Jackson, Kevin Tracy, and Zachary Manchester. Planning with attitude. IEEE Robotics and Automation Letters, 6(3):5658–5664, 2021.

[6]

Oussama Kanoun, Florent Lamiraux, and Pierre-Brice Wieber. Kinematic control of redundant manipulators: generalizing the task-priority framework to inequality task. IEEE Transactions on Robotics, 27(4):785–792, 2011.

Software References

MJINX builds upon and integrates with the following software libraries:

  • JAX: Autograd and XLA for high-performance machine learning research.

  • MuJoCo: Physics engine for detailed, efficient robot simulation.

  • PINK: Differentiable inverse kinematics using Pinocchio.

  • MINK: MuJoCo-based inverse kinematics.

  • Optax: Gradient processing and optimization.

  • JaxLie: JAX library for Lie groups.

Acknowledgements

MJINX would not exist without the contributions and inspiration from several sources:

  • Simeon Nedelchev for guidance and contributions during development

  • Stéphane Caron and Kevin Zakka, whose work on PINK and MINK respectively provided significant inspiration

  • The MuJoCo MJX team for their excellent physics simulation tools

  • IRIS lab at KAIST

Citing MJINX

If you use MJINX in your research, please cite it as follows:

@software{mjinx25,
author = {Domrachev, Ivan and Nedelchev, Simeon},
license = {MIT},
month = mar,
title = {{MJINX: Differentiable GPU-accelerated inverse kinematics in JAX}},
url = {https://github.com/based-robotics/mjinx},
version = {0.1.1},
year = {2025}
}