MULTI-FINGER PREHENSION: BIOMECHANICS AND CONTROL

  • Vladimir M. Zatsiorsky
Keywords: prehension, grasping, finger forces, enslaving, neural network, motor redundancy

Abstract

Since 1998 our group published about 20 papers in peer- reviewed journals on biomechanics and control of multi-finger tasks (see a Reference List). The research was done together with Dr. M.L. Latash in cooperation with post-doctoral fellows and graduate students Dr. F. Oanion, Z.-M. Li, S.Li, R. Gregory, F.Gao and T.Pataky. The goal of this presentation is to review some of these publications and to report on new results. Many sports-from basketball to javelin throwing and from archery to racket sports-require grasping and manipulation of hand-held objects. Study of multi-finger prehension is an imperative field of research: although human civilization has been build by hands, regrettably we know little about hand functioning. Numerous practical applications of the problem range from clinics and ergonomics to robotics. In multi-finger grasps, the fingers are statically redundant-the number of unknown forces exceeds the number of equilibrium equations-and kinematically over- onstrained, a variation in the position of a grasped object affects the position of all the fingers (likewise, a joint angle defines the length of all the muscles crossing the joint). The grasping hand is a convenient object to study the motor redundancy problem because all the involved forces can be directly measured and the sharing pattern easy documented. This is not available when the motor redundancy problem is addressed at the level of individual muscles and their contribution into the total joint torque-a most popular object for studying the sharing problem. Two considerations, a general and a specific one inspired this study. From a general perspective the idea is to study the problem of motor redundancy using the fingers as an expedient object. From a more specific standpoint, hand and finger function by itself is worthy of study.