Personal‐specific gait recognition based on latent orthogonal feature space

Exoskeleton has been applied in the field of medical rehabilitation and assistance. However, there are still some problems interaction between human exoskeleton, such as time delay, existence certain constraints on body, movement is hard to follow. A motion pattern recognition model based long short-term memory (LSTM) proposed, which can recognise state body. Meanwhile, orthogonalisation method integrated make personal-specific disentangling, it effectively improve generalisation ability different groups people, so effective follower exoskeleton. Compared with other traditional methods, this better performance stronger ability, significance exoskeleton algorithm.