Research on Adaptive Optimal Iterative Learning Control Based on Data Driven

doi:10.3901/JME.2021.17.207

Abstract

Abstract: The traditional optimal iterative learning control (TOILC) can effectively improve the tracking performance of the servo system, but there may be parametric perturbations during the operation of the system, whose parameters are constantly changing, resulting in poor convergence of TOILC, and thus the tracking performance of the system will be seriously deteriorated. Time-varying characteristics of the system, therefore, the nonparametric model identification combined with TOILC put forward a data driven based adaptive optimal iterative learning control (DDAOILC) algorithm, the iteration process according to the name of the input and output signal of the system model to identify the optimal iterative learning controller, so as to update the algorithm does not need to obtain accurate system model information in advance, make up for the deficiency of the TOILC; Simulation and experimental results show that DDAOILC can effectively deal with the servo system's time-varying characteristics, and it still has high tracking performance when the system has parameter perturbation.

Key words: iterative learning, convergence, servo system, identification, optimal control, data-driven

CLC Number:

YANG Liangliang, YUAN Rui, SHI Weimin, LU Wenqi. Research on Adaptive Optimal Iterative Learning Control Based on Data Driven[J]. Journal of Mechanical Engineering, 2021, 57(17): 207-216.

References

[1] ARIMOTO S,KAWAMURA S,MIYAZAKI F. Bettering operation of robots by learning[J]. Robot. Syst.,1984,1(2):123-140.
[2] KIM D I,KIM S. An iterative learning control method with application for CNC machine tools[J]. IEEE Transactions on Industry Applications,1996,32(1):66-72.
[3] MOORE K L. An observation about monotonic convergence in discrete-time,P-type iterative learning control[C]//IEEE International Symposium on Intelligent Control. IEEE,2001:45-49.
[4] LIU G,HOU Z. Iterative learning based model free adaptive control for subway trains with speed and input constraints[C]//Chinese Control Conference (CCC). 2019:2436-2441.
[5] NORRLÖF M,GUNNARSSON S. Experimental comparison of some classical iterative learning control algorithms[J]. IEEE Transactions on Robotics & Automation,2002,18(4):636-641.
[6] GE X,STEIN J L,ERSAL T. Frequency-domain analysis of robust monotonic convergence of norm-optimal iterative learning control[J]. IEEE Transactions on Control Systems Technology,2018,26(2):637-651.
[7] LIU L,DAI X,ZHANG W,et al. A data-driven model of optimal iterative learning control for distributed parameter systems[C]//International Conference on Information,Cybernetics,and Computational Social Systems (ICCSS). IEEE,2018:495-498.
[8] ZHANG Xinxin,LI Min,DING Huafeng,et al. Data-driven tuning of feedforward controller structured with infinite impulse response filter via iterative learning control[J]. IET Control Theory Applications,2019,13(8):1062-1070.
[9] BRISTOW D A,THARAYIL M,ALLEYNE A G. A survey of iterative learning control[J]. IEEE Control Systems,2006,26(3):96-114.
[10] 马航,杨俊友,袁琳. 迭代学习控制研究现状与趋势[J]. 控制工程, 2009,16(3):286-290. MA Hang,YANG Junyou,YUAN Lin. Current state and trend of iterative learning control[J]. Control Engineering,2009,16(3):286-290.
[11] 杨亮亮,胡建. 基于最优控制迭代学习算法的收敛性研究[J]. 机电工程,2018,35(4):397-401. YANG Liangliang,HU Jian. Research of convergence based on optimal control iterative learning algorithm[J]. Mechanical and Electrical Engineering,2018,35(4):397-401.
[12] 杨亮亮,王杰,王飞,等. 基于最优控制迭代学习的直线伺服系统振动抑制研究[J]. 机械工程学报,2019,55(15):217-225. YANG Liangliang,WANG Jie,WANG Fei,et al. Research on vibration suppression of linear servo system based on optimal control iterative learning[J]. Journal of Mechanical Engineering,2019,55(15):217-225.
[13] GUNNARSSON S,NORRLÖF M. Some aspects of an optimization approach to iterative learning control[C]//IEEE Conference on Decision & Control. IEEE,1999:1581-1586.
[14] GUNNARSSON S,NORRLÖF M. On the design of ILC algorithms using optimization[J]. Automatica,2001,37(12):2011-2016.
[15] DIJKSTRA B G,BOSGRA O H. Extrapolation of optimal lifted system ILC solution,with application to a wafer stage[C]//American Control Conference. IEEE,2002:2595-2600.
[16] DIJKSTRA B G,BOSGRA O H. Exploiting iterative learning control for input shaping,with application to a wafer stage[C]//American Control Conference. IEEE,2003:4811-4815.
[17] van OOSTEN C L,BOSGRA O H,DIJKSTRA B G. Reducing residual vibrations through iterative learning control,with application to a wafer stage[C]//American Control Conference. IEEE,2004:5150-5155.
[18] LIN N,CHI R,HUANG B. Linear time-varying data model-based iterative learning recursive least squares identifications for repetitive systems[J]. IEEE Access,2019,7:133304-133313.
[19] ZHOU L,JIA L,WANG Y L. Quadratic-criterion-based model predictive iterative learning control for batch processes using just-in-time-learning method[J]. IEEE Access,2019,7:113335-113344.
[20] YU Q,HOU Z,XU J X. D-type ILC based dynamic modeling and norm optimal ILC for high-speed trains[J]. IEEE Transactions on Control Systems Technology,2018,26(2):652-663.
[21] CHI R,HOU Z,HUANG B,et al. A unified data-driven design framework of optimality-based generalized iterative learning control[J]. Computers & Chemical Engineering,2015,77(9):10-23.
[22] CHI R,HOU Z,JIN S,et al. Improved data-driven optimal TILC using time-varying input signals[J]. Journal of Process Control,2014,24(12):78-85.