Examination of a Collision Detection Algorithm for Predicting Grip Posture of Small to Large Cylindrical Handles 2006-01-2328
A 3-dimensional kinematic model for predicting grip posture was developed. The finger joints are all rotated at a constant rate until contact is detected between the fingers and the work object. By comparing the model’s predicted hand postures with experimental data, it was shown that the model gave reasonable predictions (R2=0.72). The model predicts MCP (Metacarpophalangeal) and PIP (Proximal Interphalangeal) joint angles better than it predicts DIP (Distal Interphalangeal) joint angles.
A sensitivity study using this model was performed. The hand length, hand breadth, object size and skin deformation level were changed and the effects of these factors on hand posture was examined. The hand length, hand breadth and skin deformation level do not seem to affect hand posture much. But the change in object size affects hand posture much more than other factors.