Human Preference-Based Learning for High-dimensional Optimization of Exoskeleton Walking Gaits
This addresses the challenge of personalizing exoskeleton gaits for clinical and rehabilitation applications, though it is an incremental improvement over prior preference-based methods.
The paper tackles the problem of optimizing exoskeleton walking gaits for user comfort in high-dimensional parameter spaces, where existing methods are limited to low dimensions, and presents LineCoSpar, a human-in-the-loop framework that achieves sample-efficient optimization in simulations and human trials.
Optimizing lower-body exoskeleton walking gaits for user comfort requires understanding users' preferences over a 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, this work identifies gait attributes that characterize broader preferences across users. In simulations and human trials, we empirically verify that LineCoSpar is a sample-efficient approach for high-dimensional preference optimization. Our analysis of the experimental data reveals a correspondence between human preferences and objective measures of dynamicity, while also highlighting differences in the utility functions underlying individual users' gait preferences. This result has implications for exoskeleton gait synthesis, an active field with applications to clinical use and patient rehabilitation.