腱駆動ロボットの知的制御 : 位置・力ハイブリッドANN制御

概要

論文の詳細を見る
The purpose of this paper is to design force/position hybrid intelligent controllers for a tendon-deriven robot manipulates Tendon-driven robot manipulates are expected to be more suitable for force control tasks than traditional motor-driven robot manipulaters because of the light weightness, the joint stiffness adjustability, and especially the robust stablity for the unexpected contact with the environments. In this paper, combining the joint-space orthogonalization method [1] and the tendon-driven mechanism controller [2], we construct two types of controllers: a non-adaptive controller and an adaptive neural network (ANN) controller for a tendon-driven robot manipulates The stability for both contorllers are then studied. Finally, the efficiency is shown by several numerical simulations.
明治大学の論文