Abstract
Workload depends on each persons working posture and a compelled posture causes their fatigue. In order to reduce workload, working postures have been analyzed. But, now, it depends for the classification of a work posture on an analysis person's eyes, it is difficult to extract and classify posture data correctly if it is not the skilled technician. In this study, we propose a method that it is able to perform an exact posture classification by focusing the angle of joint on working personnel. We get joint angle data from motion data, we developped posture classification method using fuzzy inference based on some bending joint angle. By using fuzzy inference, we can classify vague posture which overlapped some segment of basic posture. From experiment, we found that we get the classification result of a posture high accuracy.
