Corpus ID: 212725029

Human Preference-Based Learning for High-dimensional Optimization of Exoskeleton Walking Gaits

@article{Tucker2020HumanPL,
  title={Human Preference-Based Learning for High-dimensional Optimization of Exoskeleton Walking Gaits},
  author={Maegan Tucker and Myra Cheng and Ellen R. Novoseller and Richard Cheng and Yisong Yue and Joel W. Burdick and Aaron D. Ames},
  journal={ArXiv},
  year={2020},
  volume={abs/2003.06495}
}
  • Maegan Tucker, Myra Cheng, +4 authors Aaron D. Ames
  • Published 2020
  • Engineering, Computer Science
  • ArXiv
  • Understanding users' gait preferences of a lower-body exoskeleton requires optimizing over the high-dimensional gait parameter space. However, existing preference-based learning methods have only explored low-dimensional domains due to computational limitations. To learn user preferences in high dimensions, this work presents LineCoSpar, a human-in-the-loop preference-based framework that enables optimization over many parameters by iteratively exploring one-dimensional subspaces. Additionally… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 29 REFERENCES

    A Method for Online Optimization of Lower Limb Assistive Devices with High Dimensional Parameter Spaces

    VIEW 3 EXCERPTS

    Towards Restoring Locomotion for Paraplegics: Realizing Dynamically Stable Walking on Exoskeletons

    VIEW 1 EXCERPT

    Towards Variable Assistance for Lower Body Exoskeletons

    VIEW 1 EXCERPT

    Human-Inspired Control of Bipedal Walking Robots

    • Aaron D. Ames
    • Computer Science, Mathematics
    • IEEE Transactions on Automatic Control
    • 2014
    VIEW 2 EXCERPTS