主催: 一般社団法人 日本機械学会
会議名: ロボティクス・メカトロニクス 講演会2024
開催日: 2024/05/29 - 2024/06/01
The combination of the seat-type body weight support (BWS) and active treadmill system is powerful to train a variety of gait functions such as discrete/rhythmic stepping, balance, speed, and symmetry. As a supplementary technology for robot-aided gait rehabilitation, this study aims to explore methods for estimating ankle push-off initiation focusing on pelvic and saddle motion peculiar to saddle-seat-type BWS treadmill walking. The experimental results of explored methods suggest that (1) the timing estimated from transverse pelvic rotation moderately agreed with ankle push-off initiation timing (Spearman's rank correlation coefficient ρ=0.59, p<0.01) and (2) the timing estimated from transverse saddle rotation strongly agreed with ankle push-off initiation timing (ρ=0.67, p<0.01). The utilization of a single IMU sensor requires a minimal setup for identifying ankle push-off initiation, which contributes to reduce time, efforts, and costs in clinical settings.