Gait phase recognition using deep convolutional neural network (dcnn) with imu data

Gait phase recognition is of great importance to develop accurate timing feedback for exoskeleton control. In order for a powered exoskeleton to determine and provide proper assistance to the wearer during gait, the user’s current gait phase must first be identified accurately. Deep convolutional neural networks (DCNN) is a machine learning approach that is widely used in image recognition. User kinematics, described with IMU data, can be considered as an ‘image’ since it exhibits some local ‘spatial’ pattern if we put the sensor data in sequence. We propose a specialized DCNN to distinguish 5 gait phases in a gait cycle, based on IMU data and classified with foot switch information. The DCNN showed 98% accuracy during an offline experiment of gait phase recognition.