首页> 外文会议>International conference on soft computing >Grasping Control of Robot Hand by Muti-valued Neural network and Double Octahedron Force Sensors
【24h】

Grasping Control of Robot Hand by Muti-valued Neural network and Double Octahedron Force Sensors

机译:用Muti-Valive神经网络抓取机器人手和双八面体力传感器

获取原文

摘要

In this study, Double Octahedron Force Sensor (D.O.F.S.) for a robot hand has been developed for detecting touch information which has high sensitivity and high strength. In order to control the sensor intelligently, the Multi-valued Neural Network (M.N.N.) using partially-linearized sigmoid function is also proposed. M.N.N. can give the optimum grasping force according to information from D.O.F.S.. The practical application will show that D.O.F.S. and the intelligent control by M.N.N. are very effective.
机译:在本研究中,已经开发了用于检测具有高灵敏度和高强度的触摸信息的机器人手的双八面体力传感器(D.O.F.S.)。 为了智能地控制传感器,还提出了使用部分线性化的SIGMOID功能的多值神经网络(M.N.N.)。 M.N.N. 可以根据来自D.O.F.S的信息提供最佳抓握力..实际应用程序将显示D.O.F.S. 和M.N.N的智能控制。 非常有效。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号