An application of cerebellar control model for prehension movements

Abstract

At present, it is a research hotspot in robot control field that using cerebellum control model with neurophysiology significance to control various different sensory coordination motor of robot system, and constructing arm transport balanced control model is an important subject of investigating robotics and control science. In this paper, we account for the temporal coordination problem between arm transport and hand preshape during reach and grasp tasks with the general cerebellar control model proposed by us. And it has also be suggested that how the structure could learn two key functions required in the classical Hoff–Arbib theory, namely state look-ahead and TTG (time-to-go) estimation. The simulation includes two steps. Firstly, the situation about the one-dimensional space is studied and trained. The results show that the model can obtain accurate smooth trajectories, and this part will be described in detail in this paper. Secondly, we extended the simulation to the two-dimensional space. Using the method described in reference Ruan and Zhang (Chin J Electron 35(5):991–995, 2007) which replaces the position scalar of above simulation with double joint plane arm, we transform the distance training in one dimension into multi-dimensional training (direction and distance) in Cartesian space so that increase the evaluation standard for system complexity and practical applicability. The simulation results are very ideal. But in consideration of article length, the simulation process and result of this part will be introduced and explained in another paper.