Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/1774
標題: 類神經式阻抗控制法於機械臂力量控制之應用
Neural-Impedance Control for Robots Contact Tasks
作者: 楊偉勝
Yang, Wei-Sheng
關鍵字: Neural Network;類神經網路;Impedance Control;Force Control;阻抗控制法;力量控制
出版社: 機械工程學系
摘要: 
本文裡提出一種類神經式阻抗控制法,使機械手臂與環境接觸的力量控制
中,在未知環境參數之下,也能夠達到所要求的控制結果。此控制器乃是
由一阻抗控制器與線上學習式類神經網路所組成。類神經網路的作用是用
來學習在不確定環境之下整個系統狀態與阻抗控制器之參考位置的關係,
也就是目標阻抗之二階關係式。如果輸入阻抗控制器的參考位置對於受控
系統的輸出力量有誤差存在時,便利用線上學習機構修正並調整神經網路
之權值,使控制系統能夠達到控制命令。另外,根據以上單臂機械手臂力
量控制的概念,設計雙臂機械手臂控制方法來完成挾持、搬運物件的動作
。以上所介紹的各控制設計將以電腦模擬的方式來實現,並驗證與分析模
擬結果。

A neural-impedance controller is proposed in this thesis.The
objective of this controller is to track a reference force in
the presence of unknown environment dynamics during constraint
motion. This controller consists of an impedance controller and
an online-training neural network. The function of neural
network is to learn the relation between the states and the
reference position input in the impedance system with unknown
environment, or says to learn the "Target Impedance." If it
consists errors in contact force signal, the online-training
mechanism can regulate the weight in the neural network to make
the signal fits the force command. Furthermore, the proposed
neural-impedance controlleris also applied to dual-arm robot.
Simulation results show that the proposed neural-impedance
controller has satisfactory performance.
URI: http://hdl.handle.net/11455/1774
Appears in Collections:機械工程學系所

Show full item record
 

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.