基于径向基函数网络的MOTOMAN机械手运动学逆解

从集合和数学观点 ,把运动学正解和逆解问题作为机器人关节空间和工作空间之间的非线性映射关系 ,将运动学逆解过程转换为神经网络权值训练问题。基于具有局部逼近能力的特点 ,将正解结果作为训练样本 ,用 6输入、单输出的RBF网络 ,实现了MOTOMAN机械手运动学逆解计算 ,避免了传统方法的繁琐公式推导。算例表明 ,采用RBF网络解决逆解问题比BP网络的计算精度略有提高。此外 ,RBF网络有更快的收敛速度

A RBF-Network-Based Method for Solving Inverse Kinematics of MOTOMAN Manipulator

The direct and inverse kinematics can be seen as a nonlinear mapping between the joint space and the operation space of the robot, and the inverse kinematics problem can also be transformed into the weight training problem of the neural network from the point of view of the set theory and the mathematics. Because of its local approaching ability, the RBF network of 6 inputs and 1 output was designed. Meanwhile, some forward kinematics results were used as training data set, with which the inverse kinematics result was obtained and some complicated derivation procedure was avoided. Examples are given to illustrate that RBF networks not only have better computation precision than BP networks, but also converge faster than BP networks.