A method for adjusting the complexity of reference trajectories for active training of upper limb rehabilitation robots based on Bayesian optimization

Abstract

The invention discloses a method for adjusting the complexity of reference trajectories for active training of an upper limb rehabilitation robot based on Bayesian optimization. The steps are as follows: establishing a training task model, selecting a motion performance index, determining the complexity of the task trajectory, and learning the complexity of the motion performance index and the task trajectory. The functional relationship between degree parameters and the adaptive adjustment of training tasks. Since the subject's motor learning ability may change during the training process, the present invention monitors the subject's exercise performance in each round of training, and adaptively adjusts the difficulty of the training task according to the subject's exercise performance, increasing the variability of the task. Can improve the rehabilitation training effect of the subjects. The present invention monitors the sports performance of different subjects, can realize individualized self-adaptive adjustment of training tasks for each subject, and can reduce the boring and cumbersome work of therapists / technicians adjusting task difficulty for different subjects in real time increase the efficiency of rehabilitation training.

Description

technical field [0001] The invention belongs to the field of upper limb rehabilitation robots, and in particular relates to a method for adjusting the complexity of an active training reference trajectory of an upper limb rehabilitation robot. Background technique [0002] According to statistics, more than 20% of stroke patients cannot survive due to lack of timely treatment, and more than 75% of survivors are unable to take care of themselves due to hemiplegia and motor dysfunction caused by stroke. The upper limbs of the human body play an important role in daily life activities, and the hemiplegia or motor dysfunction of the upper limbs will directly lead to the loss of self-care ability of the patients. Practice has shown that exercise therapy based on the principle of brain plasticity is the most common rehabilitation treatment for upper limb hemiplegia today. Its exercise therapy, that is, through the long-term and repeated application of a certain intensity of force...

AI Technical Summary

Problems solved by technology

However, studies have shown that constant training tasks will cause patients to generate memory and cannot achieve real motor learning, and increasing task variability / diversity can improve motor skill learning and memory

Images