Human body action acquisition method based on inertial sensor without standard posture correction

Abstract

The invention discloses a human body action acquisition method based on an inertial sensor without standard posture correction. The method comprises the steps that 1, an IMU is arranged, and joint physiological kinematics constraints are established; 2, by utilizing the IMU in the step 1, when the number of sampling points of the IMU exceeds 100, an optimization program runs according to a constraint equation, and a joint axis and a joint position vector are estimated by utilizing a Gauss-Newton method; 3, two groups of joint angles are respectively solved through acceleration information andangular velocity integration by utilizing the solved joint axis and joint position vector; and 4, weighted average of the two joint angles in the step 3 is solved through complementary filtering, andthe joint angles are solved. The method aims to decode IMU signals of lower limbs of a human body and calculate angles of hip joints, knee joints and ankle joints of the lower limbs in real time according to a kinematic model of the lower limbs of the human body.

Description

technical field [0001] The invention belongs to the technical field of greenhouses; in particular, it relates to a human motion collection method based on an inertial sensor that does not require standard posture correction. Background technique [0002] Today, people's requirements for the convenience and comfort of life and work are increasing. Therefore, using the flexible exoskeleton of the lower limbs to assist the human body during human activities, instead of the human muscle system to do work, thereby reducing the energy consumption of the human body has received more and more attention. In the research of human body assist, more and more researchers have been involved in how to optimize the assist curve, improve the assist effect and user experience of the exoskeleton. The exoskeleton assist curve is based on the joint angle curve of the human body when walking. However, in the current research, because it is difficult to obtain the joint angle information of the e...

AI Technical Summary

Problems solved by technology

This kind of method often needs to make a special calibration tool, and the operation is more complicated. When the sensor position is offset, the experiment cannot continue, and it needs to be re-calibrated to continue to calculate the joint angle.

Images