基于神经网络的机器人手臂模型辨识

Neural network based model identification of robot arm

介绍了神经网络动态建模方法 ,以非线性两关节机器人手臂为例 ,用两个并联的BP网络对其进行了辨识研究。仿真结果表明 。

A dynamic modelling of neural network is introduced. Taking two-knuckle robot arm as an example, this paper investigates the identification of the nonlinear system for robot arm by using two parallel backpropagation neural networks. Simulation results show that neural network is an effective approach to model robot arm.