A Novel Method of Pattern Recognition Based on TLSTM in lower limb exoskeleton in Many Terrains

Abstract

The lower limb exoskeleton robot is a kind of assistance equipment in human movement, which is widely applied to assist the human body to carry heavy loads and reduce the burden. The other hand, it is be used to help the disabled walk during early rehabilitation. In this paper, TLSTM method consisting of two-layer LSTM is proposed for pattern recognition in lower limb exoskeleton. The Inertial Measurement Units (IMUs) installed on the exoskeleton is applied to collect movement data, which is used to extract the features. The structure of TLSTM neural network is designed for pattern recognition, such as level ground walking, stair ascending, stair descending, ramp ascending and ramp descending. TLSTM neural network is trained with the preprocessed training set. The model of TLSTM neural network is tested with the test set. The experimental result shows that accuracy of TLSTM can reach 97%. Compared with LSTM, BP and SVM algorithms, TLSTM has an excellent effect in pattern recognition in many terrains. Pattern recognition paves the way for compliant control of the lower limb exoskeleton robot.