A geometric algorithm to predict the arm reach posture for computer-aided ergonomic evaluation
WANG ; VERRIEST
Type de document
ARTICLE DE PERIODIQUE
Langue
anglais
Auteur
WANG ; VERRIEST
Résumé / Abstract
Simulation of task-oriented human postures is one of the essential functions of a computerized human model for making reach, vision and fit analyses in a computer-aided design environment. After reviewing different existing methods of posture prediction, a geometric inverse kinematic algorithm to predict arm reach postures has been proposed, based on the criterion of minimization of the norm of joint angular velocities. The arm is modelled as a four-degrees-of-freedom kinematic linkage system, three for the shoulder and one for the elbow. A detailed three-dimensional description of the shoulder joint motion range is given. The main advantage of the proposed method is that it can take into account the non-linearity of the shoulder joint limit in a direct and easy way. Another important advantage is that no matrix inverse calculation is needed, thus reducing the calculation time. Experimental validation shows that the arm reach postures predicted by the proposed method are very close to the real ones in a large arm-reachable space. If initial arm postures are not too awkward, no additional manipulation is needed to correct the predicted arm reach posture from a visual criterion. The proposed method of arm posture prediction can be used as an efficient arm posture manipulation primitive.
