Gait Event Detection of a Lower Extremity Exoskeleton Robot by an Intelligent IMU

• Published in: IEEE sensors journal Vol. 18; no. 23; pp. 9728 - 9735
• Main Authors: Ding, Shuo, Ouyang, Xiaoping, Liu, Tao, Li, Zhihao, Yang, Huayong
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.12.2018; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Acceleration; accuracy and detection rate; Angular velocity; Data processing; Data sampling; Exoskeletons; Fitness equipment; Foot; foot angular velocity and acceleration; Force plates; Gait; gait event detection; Inertial platforms; Intelligent inertial measurement unit; Legged locomotion; lower extremity exoskeleton robot; Robot control; Robots; Sensors; Time lag; Walking; Windows (intervals)
• ISSN: 1530-437X; 1558-1748
• DOI: 10.1109/JSEN.2018.2871328

To control an exoskeleton robot, gait event (phase) of the robot needs to be identified. This paper presents a novel gait event detection method for a lower extremity exoskeleton robot based on an intelligent inertial measurement unit (iIMU). The iIMU is designed as a gait monitor to independently accomplish data sampling, data processing, and wireless transmission. It also has good portability that can be easily attached on the surface of a shoe. Moreover, an online detection algorithm is proposed to detect the gait events by local search windows and fixed thresholds, resulting in a minimal time delay and small computational burden. The accuracy and detection rate of the iIMU are experimentally verified by 10 healthy subjects walking on a force plate and treadmill. The mean time errors of heel-strike and toe-off detection are -10 and 19 ms when compared to the force plate. Gait events of a total 478 steps, collected from a treadmill with various walking speeds, are all detected. When applied to a lower extremity exoskeleton robot, the iIMU successfully detects the gait events of the human-robot synchronous walk.