This paper presents a control model for object manipulation. Properties of objects and environmental conditions influence the\r\nmotor control and learning. System dynamics depend on an unobserved external context, for example, work load of a robot\r\nmanipulator. The dynamics of a robot arm change as it manipulates objects with different physical properties, for example, the\r\nmass, shape, or mass distribution. We address active sensing strategies to acquire object dynamical models with a radial basis\r\nfunction neural network (RBF). Experiments are done using a real robot�s arm, and trajectory data are gathered during various\r\ntrials manipulating different objects. Biped robots do not have high force joint servos and the control system hardly compensates\r\nall the inertia variation of the adjacent joints and disturbance torque on dynamic gait control. In order to achieve smoother control\r\nand lead to more reliable sensorimotor complexes, we evaluate and compare a sparse velocity-driven versus a dense position-driven\r\ncontrol scheme.
Loading....