Multi-objective optimisation approach for predicting seated posture considering balance

Common tasks of vehicle operation are performed while seated. Using a whole-body human articulated model, we propose a methodology of predicting realistic seated postures that is based on optimising multiple-objective functions. The idea of zero-moment point (ZMP), inspired from the stability of bipedal motion, is applied to the seated postures as the main criterion for stability. An efficient formulation of ZMP is derived from the resultant reaction loads, which are also necessary to predict seat reaction forces. The required joint actuator torques is calculated and is within the limits. Several seated posture examples with different tasks (external forces and target points) show associated natural postures with different stability levels.