The field of nonlinear control has demonstrated success towards
realizing complex robotic behaviors, including bipedal locomotion.
Most importantly, nonlinear controllers account for the full system dynamics,
allowing for theoretical guarantees. However, when a human is introduced into the system,
our knowledge of the system dynamics decreases. Thus, a critical step to maintaining theoretic
guarantees in cases when a human is part of the system, such as the case with lower-body assistive
devices, is to better define the theoretical conditions underlying provably robust and stable locomotion.
Once we better understand these notions, we can develop systematic methods of achieving robust and stable
locomotion on a variety of bipedal platforms.
Human-Robot Interaction Project
In order to realize complex robotic behaviors in the real world,
human operators often have to manually tune various parameters ranging
from low-level controller gains to high-level features. For example, the
constraints of gait generation optimization problems are often manually tuned
to obtain walking gaits that result in stable, robust, and visually-appealing
bipedal locomotion. Also, for locomotion on lower-body exoskeletons, gait features
such as step length and step duration are typically tuned until the resulting gait
is comfortable for an individual user. Thus, to speed up this process of user-customization
and enable faster realization of stable experimental locomotion, our research aims to develop
a systematic method of customization by learning directly from subjective human feedback.
Importantly, by leveraging only *subjective* human feedback,
we can take advantage of a human's natural ability to judge *good* behavior in ways that
are not able to be captured numerically.
Biomechanics Project
Ultimately, the success of robotic behaviors can only be evaluated through implementation in
the real world. For example, the success of lower-body assistive devices is evaluated through
their translation to clinical settings. Thus, our research aims to evaluate robotic-assisted locomotion in
real-world scenarios, with an emphasis on translating user-customized stable locomotion to clinical settings.
