内容简介：　　

　　When looking around our own world, we can be convinced that most engineered machines and systems are of repetitive nature and/or are used for repetitive operations. Lifts in tall buildings run up and down along the same path with the same stopping positions. Robots are used to carry out the same tasks repetitively. Machines in factory assembly lines carry out the same operations repetitively to produce large quantities of the same products. Hard-disk drives in our computers write and retrieve data in storage disks with repetitive motions. Chemical indus¬tries have many batch processes and manufacturing systems have many run-to-run processes. Trains run at fixed schedules over the same distance on a daily basis. Satellites cycle along the same orbits around the Earth a few times a day. Power generators produce periodic AC voltages and currents which are required to be matching the reference frequency and with quality waveforms fbr feeding into power grids. Power electronics devices, such as converters, are to ensure output voltages or currents sinusoidal with ininimuin distortion.
  In the past 30 years, the repetitive feature of such machines/systems has been exploited to meet the ever-increasing demand for belter precision and perfor¬mance. Theory and designs of iterative learning controllers have been developed taking advantage of the repetitive operations to improve the tracking performance and accuracy. This improvement over iterations is not achievable using feedback controllers. A majority of the published literature on iterative learning control is on time domain analysis and design methods, including a few books on this topic. This book addresses the analysis and design of iterative learning control in fre¬quency domain as well as digital implementation of iterative learning control for industrial systems and machines. One distinctive advantage of frequency domain approach is the clear view of bandwidth which should ensure learning of useful signal components but filtering out unwanted interferences and disturbances. This book offers a spectrum of analysis and design methodologies and techniques to tune the cut-oif frequency for iterative learning controllers. Pscudo-downsampled- data schemes are developed to implement iterative learning controllers with good cut-off frequency. These developed designs and techniques ensure stable learning transient and monotonic convergence performance over iterations and at the same time include as many as possible signal components in the given tracking tasks. This book is aimed for practitioners/engineers, senior undergraduate students as well as postgraduate students in control engineering.

英文目录：
1 Introduction
　　1.1 Background
　　　　1.1.1 What Is ILC?
　　　　1.1.2 A Brief History
　　1.2 Basics of ILC
　　　　1.2.1 ILC Formulation
　　　　1.2.2 Comparison of ILC in Different Domains
　　1.3 ILC Design and Analysis
　　　　1.3.1 ILC Learning Laws
　　　　1.3.2 Two ILC Configurations
　　　　1.3.3 Convergence Analysis
　　　　1.3.4 Transient Analysis
　　1.4 Robotic System with ILC
　　1.5 About the Book
　　References
2 Learnable Band Extension and Multi-channel Configuration
　　2.1 A-Type Learning Control
　　2.2 Convergence Analysis of A-Type ILC
　　2.3 Design of A-Type ILC
　　　　2.3.1 Lead-Time Selection
　　　　2.3.2 Gain Selection
　　　　2.3.3 Robustness in Design
　　2.4 A Design Example of A-Type ILC
　　　　2.4.1 Learning Control Design
　　　　2.4.2 Comparison of D-, P-, PD-, and A-Type ILCs
　　　　2.4.3 Case Study and Experiments
　　2.5 A-Type ILC Based Multiple Channel Learning
　　　　2.5.1 Multi-channel Structure for ILC
　　　　2.5.2 Error Separation
　　2.6 Multi-channel A-Type ILC
　　2.7 Design of Multi-channel A-Type ILC
　　2.8 Robot Application of Multi-channel A-Type ILCs
　　2.9 Conclusion
　　References
3 Learnable Bandwidth Extension by Auto-Tunings
　　3.1 Cutoff Frequency Tuning
　　　　3.1.1 Objective and Problems
　　　　3.1.2 Learning Stability
　　　　3.1.3 Learning Divergence
　　　　3.1.4 Cutoff Frequency Tuning
　　　　3.1.5 Termination of Tuning
　　3.2 Lead Step Tuning
　　　　3.2.1 Basis of Tuning
　　　　3.2.2 Tuning Method
　　3.3 Experiment on Auto-Tuning ILC
　　　　3.3.1 Experiment 1: A-Type ILC with and
　　　　3.3.2 Experiment 2: One-Step-Ahead ILC with and
　　　　3.3.3 Experiment 3: Tuning Lead Step with
　　3.4 Conclusion
　　References
4 Reverse Time Filtering Based ILC
　　4.1 Best Phase Lead and Generation Method for SISO ILC System
　　4.2 Learning Control Using Reversed Time Input Runs
　　　　4.2.1 Learning Law
　　　　4.2.2 Model Based Approach
　　4.3 Comparison with Other Works
　　4.4 Case Study of Robot Application
　　　　4.4.1 Exact Zero Phase
　　　　4.4.2 Reverse Time Filtering Using a Model
　　　　4.4.3 Robot Performance and Experiments
　　4.5 MIMO ILC System and Error Contraction
　　4.6 Clean System Inversion ILC
　　4.7 System Hermitian ILC
　　4.8 An Example of Robot Joints and Experiments
　　4.9 Conclusion
　　References
5 Wavelet Transform Based Frequency Tuning ILC
　　5.1 Wavelet Packet Algorithm for Error Analysis
　　　　5.1.1 Wavelet Packet Algorithm
　　　　5.1.2 Error Analysis Using Wavelet Packet Algorithm
　　5.2 Cutoff Frequency Tuning ILC
　　　　5.2.1 Cutoff Frequency Tuning Scheme
　　　　5.2.2 Design of Zero-Phase Low-Pass Filter
　　5.3 Time-Frequency Domain Analysis
　　5.4 Case Study of Frequency Tuning ILC
　　　　5.4.1 Determination of Iearning Gain
　　　　5.4.2 Determination of Lead Step
　　　　5.4.3 Determination of Decomposition Level
　　　　5.4.4 Experimental Results
　　5.5 Conclusion
　　References
6 Learning Transient Performance with Cutoff-Frequency Phase-In
　　6.1 Upper Bound of Trajectory Length for Good Learning Transient
　　6.2 Cutoff-Frequency Phase-In Method
　　6.3 Sliding Cutoff-Frequency Phase-In Method
　　6.4 Robot Case Study with Experimental Results
　　　　6.4.1 Parameter Selection
　　　　6.4.2 Overcoming Initial Position Offset
　　　　6.4.3 Improving Tracking Accuracy
　　6.5 Conclusion
　　References
7 Pseudo-Downsampled ILC
　　7.1 Downsampled Learning
　　　　7.1.1 Pseudo-Downsampled ILC
　　　　7.1.2 Two-Mode ILC
　　7.2 Learning Data Processing
　　　　7.2.1 Signal Extension
　　　　7.2.2 Anti-aliasing Filtering and Anti-imaging Filtering
　　　　7.2.3 Simulation Results
　　7.3 Convergence Analysis
　　　　7.3.1 Convergence of Pseudo-Downsampled ILC
　　　　7.3.2 Convergence Analysis of Two-Mode ILC
　　7.4 Experimental Study of Downsampled ILC
　　　　7.4.1 Parameter Selection
　　　　7.4.2 Experimental Study of Two-Mode ILC
　　7.5 Conclusion
　　References
8 Cyclic Pseudo-Downsampled ILC
　　8.1 Cyclic Pseudo-Downsampling ILC
　　8.2 Convergence and Robustness Analysis
　　8.3 Robot Application
　　　　8.3.1 Parameter Selection
　　　　8.3.2 Experiment of Cyclic Pseudo-Downsampled ILC
　　8.4 Conclusion
　　References
9 Possible Future Research
Appendix A: A Robotic Test-Bed for Iterative Learning Control