Human-like motion based on a geometrical inverse kinematics and energetic optimization

For humanoid robotics the cooperation with human beings makes human like behavior indispensable. The robots of our days have to feature human like form and structure. But, more importantly its ways of motion and reactions should be like that expected with human. In this paper, we propose an algorithm for solving the human arm inverse kinematics problem (position and orientation). The proposed method eliminates the singular configurations faced in the classical inverse kinematics methods. Our algorithm is based on the choice of a special reference frames in the workspace of the different parts of human arm. The geometrical projection in local frames gives the aimed results. The proposed redundancy resolution criterion for 7ddl human like arm is obtained through an energetic optimization. The proposed algorithm could be used for humanoid robots arm, in upper limb prosthesis application, as well as in computer animations. It gives the solution for the inverse kinematics of the arm. Simulation results within ADAMS software are presented. Our study takes the result of a detailed grasping research which gives an associated object reference frame indicating the way to grasp the object.