A Repetitive Segmented Training Neural Network Controller with Applications to Robot Visual Servoing

The authors design a neural network controller for a nonlinear system with uncertainties that are invariant or repetitive over repeatedly executed tasks. The training of the neural networks is carried out iteratively as the task repeats. The desired trajectory is segmented, and for each segment a local neural network is constructed to keep the training errors within a permitted region. Meanwhile, the training is segment-wise progressively from the starting segment to the ending one. The accurate tracking of the whole desired trajectory is thus accomplished in a step-by-step or segment-bysegment manner, which means that the training of the second segment starts after the rst segment tracking has reached a desired accuracy level. To guarantee the uniform boundedness of the pointwise training, a projection-type learning update law and deadzone technique are proposed. As an application example, a robot visual servoing control problem with an uncalibrated camera is considered. The eectiveness of the proposed neural network controller with

[1]  M. Corless,et al.  Continuous state feedback guaranteeing uniform ultimate boundedness for uncertain dynamic systems , 1981 .

[2]  Nikolaos Papanikolopoulos,et al.  Adaptive robotic visual tracking: theory and experiments , 1993, IEEE Trans. Autom. Control..

[3]  Harvey Lipkin,et al.  A dynamic quasi-Newton method for uncalibrated visual servoing , 1999, Proceedings 1999 IEEE International Conference on Robotics and Automation (Cat. No.99CH36288C).

[4]  A. Isidori,et al.  Adaptive control of linearizable systems , 1989 .

[5]  Minoru Asada,et al.  Versatile visual servoing without knowledge of true Jacobian , 1994, Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS'94).

[6]  Y. D. Landau,et al.  Adaptive control: The model reference approach , 1979, IEEE Transactions on Systems, Man, and Cybernetics.

[7]  Yih-Guang Leu,et al.  Robust adaptive fuzzy-neural controllers for uncertain nonlinear systems , 1999, IEEE Trans. Robotics Autom..

[8]  Suguru Arimoto,et al.  Bettering operation of Robots by learning , 1984, J. Field Robotics.

[9]  Marios M. Polycarpou,et al.  High-order neural network structures for identification of dynamical systems , 1995, IEEE Trans. Neural Networks.

[10]  B. Paden,et al.  Stability of learning control with disturbances and uncertain initial conditions , 1992 .

[11]  Sarangapani Jagannathan.  Discrete-time CMAC NN control of feedback linearizable nonlinear systems under a persistence of excitation , 1999, IEEE Trans. Neural Networks.

[12]  Frank L. Lewis,et al.  Feedback linearization using neural networks , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).

[13]  Rolf Unbehauen,et al.  Robot visual servoing with iterative learning control , 2002, IEEE Trans. Syst. Man Cybern. Part A.

[14]  Hideki Hashimoto,et al.  Self-organizing visual servo system based on neural networks , 1992 .

[15]  P Jiang.  STABILITY ANALYSIS FOR ITERATIVE LEARNING CONTROL AND ITS APPLICATION TO MANIPULATORS , 1997 .

[16]  James S. Albus,et al.  New Approach to Manipulator Control: The Cerebellar Model Articulation Controller (CMAC)1 , 1975 .

[17]  Frank L. Lewis,et al.  Feedback linearization using neural networks , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).

[18]  Danwei Wang,et al.  Convergence and Robustness of Discrete Time Nonlinear Systems with Iterative Learning Control , 1998, Autom..

[19]  A. Zinober,et al.  Adaptive Control: the Model Reference Approach , 1980 .

[20]  Harvey Lipkin,et al.  Uncalibrated Eye-in-Hand Visual Servoing , 2002, Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No.02CH37292).

[21]  C. S. George Lee,et al.  Adaptive image feature prediction and control for visual tracking with a hand-eye coordinated camera , 1990, IEEE Trans. Syst. Man Cybern..

[22]  Marios M. Polycarpou,et al.  Stable adaptive neural control scheme for nonlinear systems , 1996, IEEE Trans. Autom. Control..

[23]  Klaus Schulten,et al.  Implementation of self-organizing neural networks for visuo-motor control of an industrial robot , 1993, IEEE Trans. Neural Networks.

[24]  Nikolaos Papanikolopoulos,et al.  Six degree-of-freedom hand/eye visual tracking with uncertain parameters , 1994, Proceedings of the 1994 IEEE International Conference on Robotics and Automation.

[25]  Guoping Liu,et al.  Variable neural networks for adaptive control of nonlinear systems , 1999, IEEE Trans. Syst. Man Cybern. Part C.

[26]  Robert M. Sanner,et al.  Gaussian Networks for Direct Adaptive Control , 1991, 1991 American Control Conference.

[27]  Weiping Li,et al.  Applied Nonlinear Control , 1991 .

[28]  Mingxuan Sun,et al.  An iterative learning controller with initial state learning , 1999, IEEE Trans. Autom. Control..

[29]  S. Saab.  On the P-type learning control , 1994, IEEE Trans. Autom. Control..

[30]  James S. Albus,et al.  I A New Approach to Manipulator Control: The I Cerebellar Model Articulation Controller , 1975 .

[31]  Zhihua Qu,et al.  Robust Iterative Learning Control for a Class of Nonlinear Systems , 1998, Autom..

[32]  J. J. Slotine,et al.  Tracking control of non-linear systems using sliding surfaces with application to robot manipulators , 1983, 1983 American Control Conference.

[33]  Nikolaos Papanikolopoulos,et al.  Six degree-of-freedom hand/eye visual tracking with uncertain parameters , 1995, IEEE Trans. Robotics Autom..