This project contributes to the understanding of how intelligent robotic systems can use a model of human intent, perception, and behavior to enhance human-robot cooperation, as may occur during rehabilitative gait training in physical therapy. The project seeks to understand how small forces applied to the hand can be used to alter clinically-relevant human gait patterns. The project is significant because it generates the experimental data and models of human gait and sensorimotor cognition needed to develop a robotic device that will interact physically and intuitively with individuals with mobility impairments to enhance their quality of life. By developing a robotic test-bed system and the models of human sensorimotor behavior that will guide its actions, this project will serve the national interest and advance the NSF mission to promote the progress of science and to advance the national health.
The project takes a three-stage approach to advancing the objectives of the NSF's Mind, Machine and Motor Nexus (M3X) program, which are to understand how the human mind controls body movements during the manipulation of machines, and how machine response can shape and influence both the mindset and movements of the human user. First, the project team will use novel instrumented devices and motion capture technology to measure interactive human-to-human hand forces and the resulting gait motions that occur during the proposed therapeutic intervention. In Stage 2 they will use machine learning techniques to develop a model of the motor and cognitive transformations that occurred during the human-to-human experiments of Stage 1. Finally, the team will embed the models within a novel robotic test-bed that will implement physical human-robot interactions at the hands. Human-robot experiments will evaluate the ability of the test-bed to promote desired changes in clinically-relevant gait characteristics such as gait speed, step length and step cadence. Fundamental issues addressed by the project include: 1) identifying the relationships between hand forces, gait dynamics and perceived intents of the instructor (therapist) and student (patient) that arise during a simplified version of rehabilitative partner dance; 2) modeling those relationships so as to enable prediction of how small hand forces indirectly change human gait dynamics; and 3) developing a robotic gait coach capable of promoting desired changes in human gait dynamics. In addition, the project supports education and promotes diversity through innovative outreach activities that will engage teams of students and older adults from underrepresented communities in the conception, design and prototyping of mobility-related assistive technologies. The project outcomes may have long-term impact on the quality of life of millions of Americans with deficits of gait by developing a robotic system that can assist therapists in their efforts to improve patient fitness, mobility, and independence.
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.
