Gait Intention Estimation Based on Preliminary Motion for Complete Spinal Cord Injury Patients

本研究提出了一种步态意图算法，推断摆动速度以及检测的时刻开始摆动腿的基础上的初步运动，即。地面反作用力中心的行为，而不是检测生物电信号。预备运动是一种在人开始行走运动之前立即可检测到的自主运动。在使用配备有外骨骼机器人（混合辅助肢体：HAL），包括所提出的算法的人体模型的实验后，我们进行了现实的测试与严重的脊髓损伤患者（ASIA分类; A级）穿着步态支持系统。因此，HAL与所提出的算法支持病人的步态，同时调整摆动腿的速度。通过测试，验证了该算法的有效性和实际可行性。

This study proposes a gait intention algorithm that infers the swing speed as well as detects the moment to start the swing leg based on a preliminary motion, i.e.. behavior of the center of ground reaction forces, instead of detecting bioelectrical signals. The preliminary motion is a kind of voluntary motion that is detectable immediately before a person starts a walking motion. After the experiments using a mannequin equipped with the exoskeleton robot (Hybrid Assistive Limb: HAL) including the proposed algorithm, we conducted realistic tests with a severe spinal cord injury patient (ASIA classification; grade A) wearing the gait support system. As a result, HAL with the proposed algorithm supported the patient's gait while adjusting the speed of the swing leg. Through the tests, we demonstrated the effectiveness and practical feasibility of the algorithm.