| 引用本文: | 裴海龙 周其节 梁天培.基于神经网络的机械臂的力控制方法[J].控制理论与应用,1995,12(5):603~611.[点击复制] |
| PEI Hailong and ZHOU Qijie T. P. Leung.Force Control of Manipulators with Neural Networks[J].Control Theory and Technology,1995,12(5):603~611.[点击复制] |
|
| 基于神经网络的机械臂的力控制方法 |
| Force Control of Manipulators with Neural Networks |
| 摘要点击 1001 全文点击 699 投稿时间：1994-12-24 修订日期：1995-07-20 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 |
| 1995,12(5):603-611 |
| 中文关键词 神经网络 机械臂 力控制 |
| 英文关键词 neural networks robot force control |
| 基金项目 |
|
| 中文摘要 |
| 本文提出一种基于神经网络的力控制方法，由两个串联的神经网络构成机械臂的力控制器，其中一个网络用来学习机械臂本身的逆动力学系统，而另一网络用来学习未知的被接触环境的动力学特征，这种方法避免了困难的接触环境建模问题。一个双连杆机械臂的力控制的仿真实验描述了这种方法的有效性。 |
| 英文摘要 |
| In this paper, a new robot force control strategy based on neural network learning is proposed. The controller is composed of two neural networks in cascade. One of them is used to learn the inverse dynamics of the robot, while the other one to learn to express the unknown environment dynamics. With this method, the difficult environment modelling problems can be avoided. To indicate the efficiency of this algorithm, the force control problem of a two-link robot is engaged. Simulation results show that this straightforward method is very feasible. |