A biomimetic approach to mobility distribution for a human-like redundant arm

Abstract

In this paper the problem of planning trajectories using human-like vision information, and of solving inverse kinematics of robots by weighting joint involvement in a biomimetic fashion is considered. Movements of a redundant seven degree-of-freedom (DOF) robot manipulator, having a kinematic structure similar to that of the human arm, have been simulated. The design of the trajectory is based on theory about attentive vision in humans, and biomimetic considerations are used to infer laws to obtain human-like movements in inverse kinematics. The inverse kinematics is computed using a closed-loop algorithm with redundancy resolution where physically motivated secondary tasks are imposed to the internal motions, with the "virtual end-effectors" approach. Outputs of the IK algorithm related to drawing trajectories in the task space are compared with the positions of the joints of a human performing the same drawing task, computed from the images caught by a three-camera system