Lower Limb Joint Angle Estimation Based on Genetic Algorithm for Gait Analysis

Abstract

This paper presents a method to estimate joint angles of lower limb based on data of body posture gathered by 3D-image sensor. With aging society, gait analysis is a very important and interesting topic to detect and prevent their gait disorder at an early stage. Motion capture system is an effective measure for gait analysis with high precision measurement. However, most motion capture systems are expensive and a large measurement space is required. We therefore apply a 3D-image sensor to gait motion measurement and analysis. In this study, a method of joint angle estimation based on genetic algorithm to derive the joint angle using the model of inverse kinematics is proposed. A continuous model of generation for an adaptive search is applied to the joint angle estimation. We furthermore show some experimental results to discuss the applicability.