摘要
提出了基于最小二乘支持向量机的机器人逆运动学建模方法,阐述了基本设计思想和具体算法过程,与RBF神经网络相比,最小二乘向量机的优点在于其训练过程遵循结构风险最小化原则,不易发生过学习现象,它通过解一组线性方程组得到全局唯一最优解,其拓扑结构在训练结束时自动获得而不需要预先确定。通过对二自由度刚性机器人的仿真,结果验证了该方法的有效性和可行性。
A new approach to solve the inverse kinematics of robotic manipulators based on least squares support vector （LS-SVM） is presented. Compared with the RBF neural networks, the LS-SVM possesses prominent advantages： over fitting is unlikely to occur by employing structural risk minimization criterion, and the global optimal solution can be uniquely obtained owing to the fact that its training is performed through the solution of a set of linear equations. Also, the LS.SVM need not determine its topology in advance, which can be automatically obtained when the training process ends. The simulation results are presented for a 2-DOF rigid robotic manipulator, which validate the effectiveness and feasibility of the proposed method.
出处
《系统仿真学报》
EI
CAS
CSCD
北大核心
2006年第5期1260-1262,1266,共4页
Journal of System Simulation
基金
教育部科学技术研究项目(重大)(204181)
陕西省自然科学基金项目(2002F27)
陕西省教育厅专项科研计划项目(04JK248)
关键词
支持向量机
最小二乘支持向量机
神经网络
机器人逆运动学
RBF神经网络
squares support vector （SVM）
least squares support vector （LS-SVM）
neural networks
robotic manipulatorinverse kinematics
RBF neural networks
