Title :
Learning of inverse-dynamics and inverse-kinematics for two-link SCARA robot using neural networks
Author :
Onozato, Ken ; Maeda, Yutaka
Author_Institution :
Kansai Univ., Suita
Abstract :
In this paper, we describe position control of a SCARA robot using two neural networks. The simultaneous perturbation method is used for the learning of these neural networks. Two networks learn inverse kinematic and inverse dynamic respectively. Some results will be shown.
Keywords :
compliance control; industrial manipulators; learning (artificial intelligence); manipulator dynamics; manipulator kinematics; neurocontrollers; position control; robotic assembly; NN learning; inverse-dynamics learning; inverse-kinematics learning; neural networks; position control; selective compliance assembly robot arm; simultaneous perturbation method; two-link SCARA robot; Inverse problems; Kinematics; Layout; Neural networks; Perturbation methods; Position control; Recurrent neural networks; Robot control; Robot sensing systems; Service robots; Inverse Kinematics; Inverse dynamics; Inverse problem; Neural networks; SCARA robot; Simultaneous Perturbation;
Conference_Titel :
SICE, 2007 Annual Conference
Conference_Location :
Takamatsu
Print_ISBN :
978-4-907764-27-2
Electronic_ISBN :
978-4-907764-27-2
DOI :
10.1109/SICE.2007.4421135
