A Novel Adaptive ILC for Nonlinear Discrete-Time Systems Based on Neural Network
A neural network based adaptive iterative learning control is presented for a class of nonlinear discrete-time systems, in which a dead-zone scheme is introduced to enhance the robustness of the control system and to achieve arbitrary tracking accuracy. The new control approach overcomes the limitations of traditional ILC in that the target trajectory and the initial condition need not be identical when the control process repeats. Convergence analysis indicates that with the random initial states and the iteration-varying trajectories, the tracking error can converge to a bounded ball, whose size is determined by the dead-zone nonlinearity. Computer simulations verify the theoretical results.
Ronghu Chi Shulin Sui Zhongsheng Hou Wenlong Yao
School of Automation and Electrical Engineering Qingdao University of Science and Technology Qingdao School of Electronics and Information Engineering Beijing Jiaotong University Beijing, 100044 ,P.R.C School of Automation and Electrical Engineering Qingdao University of Science and Technology Qingdao
国际会议
南宁
英文
2007-07-20（万方平台首次上网日期，不代表论文的发表时间）