作 者: ;
机构地区: 南京航空航天大学机电学院
出 处: 《南京航空航天大学学报》 1997年第1期46-50,共5页
摘 要: 首次实现自组织神经网络求解机器人姿态逆解,突破了文献局限于研究位置逆解的状况。应用本文创新的自组织神经网络训练方法,结合工业机器人运动学特性,建立了工业机器人姿态逆解的神经网络方法,对PUMA560机器人的计算机仿真结果表明,该方法姿态控制精度高。 A methodology is presented for the first time whereby the self-organizing neural network is used to learn the inverse kinematic relationship between the joint-variable space and the orintation of the end-effector of a robot arm. The learning algorithm of the neural network for the arm solution of the orintation is based on the improved self-organizing neural network proposed and the study of the characters of the kinematics of a industrial robot arm. The computer simulation results for the PUMA 560 robot arm show that this method is the best in precision control of the orintation of a robot arm.
领 域: [自动化与计算机技术] [自动化与计算机技术]