Modeling Human Motor Learning, Perception, and Skill Acquisition for Movement-Assistive Robotic Technologies

This award supports research aimed at enhancing our understanding of human motor learning to improve health outcomes. The project focuses on developing movement-assistive robotic technologies to aid in human motor learning and rehabilitation, especially for complex whole-body tasks like walking. The objective is to explore how humans learn intricate balance tasks and how robotic devices can facilitate that learning process. This research looks to contribute to foundational knowledge that scientists and engineers rely on to create robotic devices capable of transforming movement rehabilitation, improving learning outcomes, and advancing our understanding of human-robot interaction. Movement-assistive robotic technologies, such as lower-limb exoskeletons, hold substantial promise for enhancing human motor learning and rehabilitation, particularly in situations where there is a perceived risk of falling. The mathematical complexities involved in modeling the complex dynamics of human movement, like walking, have hindered research into how assistive devices can aid human motor learning. This project aims to simplify the study of balance and forward propulsion by using a single-wheel vehicle model. The foundational knowledge gained looks to help determine how to apply robotic assistance to more common balance tasks, such as walking. Additionally, the study will explore the development of adaptive robotic controllers that can accelerate skill acquisition. While there has been some exploration of human motor learning for simple, two-dimensional tasks like point-to-point reaching, much less is understood about how people learn three-dimensional tasks with a high risk of falling, such as walking. Furthermore, there is limited knowledge on how robotic assistive devices should intervene to facilitate accelerated motor learning. The team will simulate paired "student" and "coach" robotic learning agents, where the student learns while the coach guides the learning process. Finally, they look to develop a robotic assistance testbed and apply insights from experiments on human learning and robot simulations to evaluate the impact of human physiological and cognitive reactions, such as fear of falling, on the mechanics of learning. The team will also test how various assistive controllers aid human balance learning. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.