A computerized dynamic biomechanical simulation model for sagittal plane lifting activities
Lin, Chiuhsiang Joe
MetadataShow full item record
The biomechanical approach provides estimation of various mechanical stresses acting on the body while a person manually handles an object. Today, photographical motion systems are available for dynamic biomechanical research; however, studies using such systems are mostly performed in the laboratory due to high cost of the equipment and the expertise required in using them. Industrial ergonomists do not have much access to dynamic analyses. To provide ergonomists with a computerized dynamic job analysis tool, a sagittal plane lifting biomechanical simulation model was developed. The model simulates the dynamic motion of lifting tasks for the five body joints: the elbow, shoulder, hip, knee, and ankle. The inputs of the model include initial and final joint postures; subject's sex, weight, and height; weight of load; lifting height; and box dimensions. On output, the angular trajectories of the five joints are predicted. The motion of the lift is completely predicted by the model without any video inputs. The simulation model contains three computation units, the dynamics computation, the trajectory formation, and the optimization unit. The dynamics unit calculates the kinematics and kinetics. The trajectory formation unit uses numerical techniques to generate smooth joint trajectories. The optimization unit consists of an objective function and several constraints that describe the behavior of the human lifting. The simulation model was validated using actual lifting data from five male and five female subjects. The lifting tasks included floor-to-knuckle and floor-to-shoulder height lifts. Comparisons were made between the actual lifting motion and predicted motion. The results showed that the simulated motion followed the actual motion closely Sensitivity analysis showed that the optimization constraints had significant effects on the prediction performance of the simulation model.