Model based control and MFAC, which is better in simulation? * *Resrach supported by NSFC under granted No. 61120106009. Hou Zhongsheng is the corresponding author.

Abstract This work presents a new data driven model free adaptive controller by virtue of the gradient information of the available plant model. Our approach is novel in the sense that we use the measured data to directly design the controller, and predict the system output using the plant model if the plant model is available. Whereas, the traditional model based control approach uses the measured data to identify the plant model firstly, and then uses the model to design and analyze the controller design based on this identified model. The main contribution is that this novel scheme opens a way for the data driven control method utilizing the plant model if the plant model is available. The theoretical analysis and the simulation results demonstrate the effectiveness and superiorities of the proposed method.

[1]  Zhongsheng Hou,et al.  Controller compact form dynamic linearization based model free adaptive control , 2012, 2012 IEEE 51st IEEE Conference on Decision and Control (CDC).

[2]  Mauro Birattari,et al.  From Linearization to Lazy Learning: A Survey of Divide-and-Conquer Techniques for Nonlinear Control (Invited Paper) , 2005 .

[3]  Håkan Hjalmarsson,et al.  From experiment design to closed-loop control , 2005, Autom..

[4]  Brian D. O. Anderson,et al.  Challenges of adaptive control-past, permanent and future , 2008, Annu. Rev. Control..

[5]  Antonio Sala.  Integrating virtual reference feedback tuning into a unified closed-loop identification framework , 2007, Autom..

[6]  Zhuo Wang,et al.  From model-based control to data-driven control: Survey, classification and perspective , 2013, Inf. Sci..

[7]  Michael Athans,et al.  Robustness of continuous-time adaptive control algorithms in the presence of unmodeled dynamics , 1985 .

[8]  R. E. Kalman,et al.  Contributions to the Theory of Optimal Control , 1960 .

[9]  Shangtai Jin,et al.  Data-Driven Model-Free Adaptive Control for a Class of MIMO Nonlinear Discrete-Time Systems , 2011, IEEE Transactions on Neural Networks.

[10]  L. Valavani,et al.  Robustness of adaptive control algorithms in the presence of unmodeled dynamics , 1982, 1982 21st IEEE Conference on Decision and Control.

[11]  Z. Hou,et al.  The model-free learning adaptive control of a class of SISO nonlinear systems , 1997, Proceedings of the 1997 American Control Conference (Cat. No.97CH36041).

[12]  R. E. Kalman,et al.  A New Approach to Linear Filtering and Prediction Problems , 2002 .