Constraint-Based Real-Time Full-Body Motion-Capture Using Inertial Measurement Units

PubDate: January 2019

Teams: National Institute of Advanced Industrial Science and Technology

Writers: Tsubasa Maruyama; Mitsunori Tada; Akira Sawatome; Yui Endo

PDF: Constraint-Based Real-Time Full-Body Motion-Capture Using Inertial Measurement Units

Abstract

Human motion measurement plays an important role in the field of motion analysis, virtual reality and ergonomics design. Recently, it has become possible to measure the human motion using a set of inertial measurement units (IMUs). However, the IMU-based motion-capture (MoCap) system only consider the ground contact constraints, so the constraints between the human and the product are not considered. Therefore, this study aims to develop a full-body motion estimation system using the IMUs, which can reflect the contact conditions between the product and the product users based on the user-specified contact constraints. In addition, the proposed system is validated through the measurement experiments of walking and chair-sitting motions. The walking experiment shows that the averaged errors of the joint angle estimation are ranged from 2.79±2.00 deg for the right ankle to 9.55±6.04 deg for the right hip. In addition, the chair-sitting experiment shows that the proposed system could estimate the sitting motion reflecting the contact condition between the subject and the chair surfaces. Our results indicate the potentials and challenges for measuring a variety of human motion using the proposed system.