Exoskeleton-oriented gait phase classification method based on TCN-HMM

Abstract

The invention discloses an exoskeleton-oriented gait phase classification method based on a TCN-HMM. The method comprises the following specific steps: 1, acquiring IMU data; 2, preprocessing the gaitdata acquired in the step 1; 3, constructing a training set and a test set; 4, constructing a hybrid TCN-HMM model, 5, training the hybrid TCNHMM model by using the training set, and 6, classifying the new walking IMU data by using the trained hybrid TCNHMM model. According to the method, the posterior probability of the state is obtained by innovatively utilizing the TCN network, and the emission probability required by the HMM model is obtained by utilizing the posterior probability, so that the hybrid TCNHMM model is formed, and the model organically combines the time characteristics and the space characteristics of the motion data together and discriminates the gait phase information. According to the method, a gait phase classification result with very high accuracy is obtained, andmeanwhile, wrong classification is inhibited.

Description

technical field [0001] The invention belongs to the field of lower limb exoskeleton human-machine collaborative motion control, and relates to a human walking gait phase classification method based on a hybrid TCN (Temporal Convolutional Networks) and HMM (Hidden Markov Model) model. Background technique [0002] In recent years, exoskeleton robots have become emerging technologies in medical, life, industrial and military applications. Among them, the lower extremity exoskeleton has shown great value. It perfectly combines human intelligence and the "physical strength" of robots, and has broad application prospects in the fields of assisting, assisting the elderly, assisting the disabled, and the military. [0003] Gait phase classification is a general method for analyzing walking motion, and accurate classification of different gait phases is crucial for controlling lower extremity exoskeletons and detecting user intent. Current gait phase recognition methods can general...

AI Technical Summary

Problems solved by technology

At the same time, in order to solve the problems of easy wear and interference in many current human body motion signal acquisition devices, the present invention adopts a more portable, durable and reliable IMU sensor, which can be installed on different parts of the body according to needs, so as to obtain Abundant human motion information

Images