The cost of therapy is one of the most significant barriers to recovery after neurological injury. Robotic gait trainers move the legs through repetitive, natural motions imitating gait. Recent meta-analyses conclude that such training improves walking function in neurologically impaired individuals. While robotic gait trainers promise to reduce the physical burden on therapists and allow greater patient throughput, they are prohibitively costly. Our novel approach is to design a new single degree-of-freedom (DoF) robotic trainer that maintains the key advantages of the expensive trainers but with a simplified design to reduce cost. Our primary design challenge is translating the motion of a single actuator to an array of natural gait trajectories. We address this with an eight-link Jansen mechanism that matches a generalized gait trajectory. We then optimize the mechanism to match different trajectories through link length adjustment based on nine different gait patterns obtained from gait database of 113 healthy individuals. To physically validate the range in gait patterns produced by the simulation, we tested kinematic accuracy on a motorized wooden proof-of-concept of the gait trainer. The simulation and experimental results suggested that an adjustment of two links can reasonably fit a wide range of gait patterns under typical within-subject variance. We conclude that this design could provide the basis for a low-cost, patient-based electromechanical gait trainer for neurorecovery.
Design of a Single Degree-of-Freedom, Adaptable Electromechanical Gait Trainer for People With Neurological Injury
Contributed by the Mechanisms and Robotics Committee of ASME for publication in the JOURNAL OF MECHANISMS AND ROBOTICS. Manuscript received August 17, 2017; final manuscript received March 21, 2018; published online May 31, 2018. Assoc. Editor: Veronica J. Santos.
Yul Shin, S., Deshpande, A. D., and Sulzer, J. (May 31, 2018). "Design of a Single Degree-of-Freedom, Adaptable Electromechanical Gait Trainer for People With Neurological Injury." ASME. J. Mechanisms Robotics. August 2018; 10(4): 044503. https://doi.org/10.1115/1.4039973
Download citation file: