Trunk and arm somatosensory measurement condition setting and adjusting device

Abstract

The invention discloses a trunk and arm somatosensory measurement condition setting and adjusting device. The trunk and arm somatosensory measurement condition setting and adjusting device comprises a base frame with adjustable position on the ground, a plurality of upright columns connected with the base frame in a horizontally adjustable manner; a plurality of handles held by a user; a plurality of pull ropes matched with the handles, wherein the handles are connected with the pull ropes, the two ends of each pull rope are connected with rope end connecting assemblies, and the rope end connecting assemblies are connected with the columns; a pair of L-shaped rotary adjusting shafts, wherein each L-shaped rotary adjusting shaft is provided with a vertical part fixedly connected with the base frame and a horizontal part capable of rotating up and down for adjustment, and the two rotary adjusting shafts are oppositely arranged left and right; a pedal structure used for supporting feet of a human body; wherein the pedal structure comprises a supporting plate, a rotating disc and a pair of pedal plates for supporting the left foot and the right foot of the human body respectively, and the rotating disc is detachably connected with and supports the pedal plates. The trunk tasks can be finely expressed and adjusted, and conditions are provided for trunk muscle cooperation behavior monitoring, somatosensory quantitative recognition and cross-experimental object trunk behavior labeling.

Description

technical field [0001] The invention relates to the technical field of human body somatosensory recognition, in particular to a device for setting and adjusting conditions for somatosensory measurement of trunk and arm. Background technique [0002] The somatosensory of the trunk muscles is the most important category of human somatosensory, but the control method of the trunk muscles is unstable, and will be constantly switched due to the influence of posture, load, balance conditions, and hand operations. To fully quantify the behavior and state of the trunk muscles, it is necessary to encode and embed the multi-dimensional sEMG signals of the trunk muscle groups into comparable vectors based on myoelectric measurement and clustering, long-short-term neural network (LSTM) and other technologies, and use RNN, Algorithmic tools such as LSTM and Attention complete action annotation. In this way, visible or invisible trunk muscle behavior can be expressed quantitatively by ve...

AI Technical Summary

Problems solved by technology

[0003] However, commercial finite element software such as ANYBODY predicts the behavior of trunk muscles based on inverse dynamics and correction of special phenomena of the human body, and cannot recognize the influence of hand task differences and dynamic balance conditions

Subtle gesture control intention and precision expectations of the hand are highly correlated with subjective purpose, and it is difficult to determine its intention based on motion monitoring, and its influence can only be determined through experiments

Images