Abstract
Human posture perception technology is still depended on traditional rigid materials, which are generally inconvenient to wear and uncomfortable and inflexible in motion. In order to solve the wearing problem and achieve long-term dynamic monitoring of human motion posture, this paper investigate the method of measuring human elbow and knee joint activity angle by electronic fabric material, and hybrid inertial sensors to perceive the whole body’s human posture. Fabric properties and the relationship between skin deformation and joint angles are characterized by combining material science, computer science, and biomechanical models. Joint angles data measured by a commercial inertial motion capture device is used as a benchmark and a BP neural network regression model is used to characterize the relationship between fabric resistance and joint angles. The experimental results show that the average absolute errors of elbow and knee angles reach 3.04° and 2.68°, respectively, compared with commercial inertial sensors; finally, the whole body posture of the human body was reconstructed in real time on the Unity 3D platform using the transmitted data from fabric sensors and inertial sensors.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have