다목적 최적화 기법을 이용한 디지털 인체 모델의 자세 예측

Abstract

Recently, in the automobile industry, user-friendly design is being emphasized for the better ergonomic design of human space (e.g., car compartments, car seat). Therefore, the test with real person is performed but it is time consuming and takes much cost. For this reason, studies trying to make the design by predicting posture and motion through the digital human model are performed. In this paper, we propose the posture prediction method of digital human model using a multi-objective optimization method . We first create a digital human model by using the anthropometric data of Korean and statistical methods. We then define constraints and objective functions that can explain the human postures and then predict the posture by solving the optimization problem.