An optimization approach to determine manual lifting motion

The dynamic behavior of a musculoskeletal link system in manual lifting is simulated by a mathematical model which contains a non-linear objective function and a set of linear, as well as non-linear constraints. The model was developed based upon the hypothesis that an individual performs the lifting motion following the principle of minimizing mechanical work done. The input data of the model includes (1) anthropometric parameters of the individual, (2) characteristics of the container (size, weight), (3) initial and final position angles of five major joints, and (4) the task performance time. The output data includes the time history of the five joint angles and the time history of five joint moments. The trajectory is such that it minimizes the mechanical work done while maximizing the utilization of all joints and remaining within the feasible region of the proposed constraints.

The simulation results indicated that the differences between the predicted motion and the measured motion is biomechanically feasible and the accuracy is adequate enough with an average U statistics ranging from 0.012 to 0.209.