Human posture prediction can often be formulated as a non-linear multi-objective optimization (MOO) problem. The joint displacement function is considered as a benchmark of human performance measures. The hypothesis is that human performance measures govern how human move as they do. Therefore, when joint displacement is used as the objective function, posture prediction is a MOO problem. The weighted sum method is commonly used to fmd a Pareto solution of this MOO problem. Within the joint displacement function, the relative value of the weights represents the relative importance of the joint. Usually, weights are determined by trial and error approaches. This paper proposes a systematic approach to determine the weights for the joint displacement function in optimization-based posture prediction where a realistic posture is given. It can be formulated as a two-level optimization problem. The design variables are joint angles and weights. The cost function is the summation of the differences between joint angles and a realistic posture. Constraints include (1) normalized weights within limits; (2) an inner optimization problem to solve for joint angles where joint displacement is the objective function, and constraints include that the end-effector reaches the target point and joint angles are within their limits. Additional constraints such as weight limits and weight linear equality constraints obtained through observations are also implemented in the formulation to test the method. A 21 degree of freedom (DOF) upper human model and three target points in-vehicle are used to illustrate the procedure of the method.
- Posture prediction
- Weighted sum