Gait trajectory generation is a crucial point in the lower limb exoskeleton. Personalized gait trajectories can be supplied to exoskeleton for different patients, because of the availability of the gait synergy. In this paper, a gait synergy model between lower limbs is established based on the Long Short-Term Memory (LSTM). The gait synergy model predicts the values of each joint on the affected side according to the angles and angular velocities of each joint on the healthy side. Three indexes are adopted to evaluate the prediction results with the R2 coefficient being more appropriate for the synergy model than the Main Absolute Percentage Error (MAPE) and Root Mean Square Error(RMSE). Experimental findings reveal that the lower limb gait synergy model based on LSTM is capable of predicting results for hip and knee joints accurately.
Gait Synergy Modeling and Joint Angle Prediction Based on LSTM
2023-08-11
4827169 byte
Conference paper
Electronic Resource
English
Efficiency of deep neural networks for joint angle modeling in digital gait assessment
BASE | 2021
|Efficiency of deep neural networks for joint angle modeling in digital gait assessment
BASE | 2021
|LSTM LSTM-based future threat prediction method and apparatus
European Patent Office | 2021
|