Action assisting device based on surface myoelectricity and action intention identification method

Abstract

The invention relates to the technical field of medical rehabilitation instruments, in particular to an action assisting device based on surface myoelectricity and an action intention identification method. A control circuit board receives a bioelectricity signal transmitted by a sensing unit and performs bioelectricity signal feature extraction; and extracted bioelectricity signal features are sent to a classifier for linear discriminant analysis, a classification result is sent to a voter, the result of the voter is compared with a threshold comparison result, a final motion intention result is output and an instruction is transmitted to a driver, so that the driver generates a driving force to control a hand exoskeleton to act. Two actions of stretching and bending are achieved by one motor, the pay-off speed is adjusted in a small range, the self-locking return stroke gap is eliminated, the safety and reliability of a self-locking mechanism are greatly improved, the difficulty of inaccurate electric signal recognition motion intention is overcome, the control delay of the whole system is reduced, and the action intention of a user can be directly reflected.

Description

technical field [0001] The invention relates to the technical field of medical rehabilitation equipment, in particular to an action assisting device and an action intention recognition method based on surface electromyography. Background technique [0002] Movement assistance equipment is mainly aimed at people with limb movement disorders, and has been widely used in neurological rehabilitation, medical and military fields. Since the use of hands is extremely important and frequent in daily life, the design of assistive devices for hand movements is particularly important. [0003] The patients with hand dysfunction can be mainly classified into two categories, one is caused by damage to the brain and neural pathways, and the other is caused by muscle damage. For the first type of patients, it can be treated by nerve transfer surgery; for the second type of patients, it can be treated by targeted muscle nerve transplantation. However, whether it is nerve transfer or targe...

AI Technical Summary

Problems solved by technology

The collection of mechanical information is convenient and the signal is stable, but it can only be obtained after the user starts exercising, and there is a serious lag. At the same time, some small hand movements cannot collect mechanical information, and the mechanical information cannot directly reflect the user's movement. intention

[0006] Motion intention recognition based on bioelectrical signals can solve the problem of hysteresis very well, but since the electromyography signal will have interference signals (such as superimposed signals, noise signals, etc.), the accuracy of recognition is low, which will affect the judgment of motion intention to mislead

Images