Dynamic simulation of hand-forearm system

We present an anthropomorphic model of a human hand, whose structural and functional properties permit the simulation of grasping tasks. The proposed model has 27-DOFs which are split into 23-DOFs for the hand, 2-DOFs for the flexion-extension and the abduction-adduction of the wrist ,and 2-DOFs for the pronation-supination and the flexion-extension of the forearm. This model allow one to simulate the grasp of an object with hand adapting to its shape by including the movements of fourth and fifth meta-carpals. The model, designed with ADAMS, is based on bio-mimetic approach which attempts to simulate the most realistic behavior during grasping tasks. In order to achieve dynamic simulations, different link masses and inertia parameters are chosen according to the human hand-forearm properties. A simulated grasping tasks is carried out using the kinematic structure, hand/grasped object contact law and position/torque control strategies. The simulation model takes into account the joint motion coupling laws and the contact distribution. Two grasping tasks are given to demonstrate the model capabilities.