The ability to walk is essential for health and well-being, particularly during childhood development. Children with cerebral palsy have neuromuscular impairments that often diminish walking capacity. Wearable exoskeletons may offer a compelling compliment to surgery and standard physical therapy. However, to be effective, exoskeleton assisted rehabilitation must be finely tailored to each individual. By seeking to improve the effectiveness of exoskeleton-assisted gait training, knowledge gained from this proposal will provide the basis for future exoskeleton intervention studies that will seek to improve underlying gait function and increased levels of habitual physical activity in cerebral palsy. The outcomes of this research have the potential to transform the treatment of pediatric gait disorders, lead to improved health, and reduce the economic burden for individuals with cerebral palsy. In addition to the scientific and societal impacts, this project seeks to broaden research participation from individuals underrepresented in STEM fields by offering interdisciplinary training to a diverse cohort of students. This research will positively impact Northern Arizona University by elevating the research profile of the Center of Bioengineering Innovation and Bioengineering PhD program.
Machine learning techniques have begun to demonstrate potential in improving human-wearable robot interactions but investigations to date have focused on single joints and short durations. The purpose of this proposal is to utilize computational intelligence to establish a framework suitable for optimizing patient-specific assistance and enhancing neuro-muscular participation during training with wearable exoskeletons. Investigators will implement, in an exoskeleton platform, an optimization framework that can learn from biomechanical responses to exoskeleton assistance to quickly converge on an optimal control strategy for the timing and magnitude of assistive joint torques. The first goal is to establish the optimization algorithms and parameters of an exoskeleton control algorithm that suitably adapts knee and ankle assistance based on walking performance. The second goal is to conduct a pilot study to determine how knee and ankle exoskeleton assistance, optimized for improving lower-extremity posture, affects gait biomechanics and neuromuscular control during consecutive daily gait training in children with cerebral palsy. It is anticipated that this research will lead to an improved understanding of the algorithm parameters needed to account for stride-to-stride and session-to-session variability for longitudinal optimization of exoskeleton control for gait rehabilitation.
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.
