Recurrent neural network modeling and learning control of flexible plates by nonlinear handling system

Proposes a trajectory control method for a flexible plates handling system with unknown parameters and joint friction. First a recurrent neural network (RNN) learns the dynamics model of the flexible plate handled by a robotic manipulator. Next, the authors obtain the feedfoward control input based on the RNN model using the proposed learning control method. The authors applied this repetitive method to both linear system and nonlinear system control. Coulomb friction is considered at the joint as the nonlinear effect. Simulation examples are conducted to show effectiveness of the proposed method.<<ETX>>

[1]  Kuldip S. Rattan,et al.  Feedforward control of flexible manipulators , 1992, Proceedings 1992 IEEE International Conference on Robotics and Automation.

[2]  Michael I. Jordan,et al.  Generic constraints on underspecified target trajectories , 1989, International 1989 Joint Conference on Neural Networks.

[3]  Bruno Siciliano,et al.  Trajectory control of a non-linear one-link flexible arm , 1989 .

[4]  Fumihito Arai,et al.  Asymptotic convergence of feedback error learning method considering spillover in controlling flexible structure , 1993, Proceedings of 1993 International Conference on Neural Networks (IJCNN-93-Nagoya, Japan).

[5]  Michael I. Jordan.  Supervised learning and systems with excess degrees of freedom , 1988 .

[6]  Yuan F. Zheng,et al.  Vibration-free movement of deformable beams by robot manipulator , 1993, [1993] Proceedings IEEE International Conference on Robotics and Automation.

[7]  Fumihito Arai,et al.  Asymptotic convergence of feedback error learning method and improvement of learning speed , 1993, Proceedings of 1993 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS '93).

[8]  Fumihito Arai,et al.  Trajectory control of flexible plate using neural network , 1993, [1993] Proceedings IEEE International Conference on Robotics and Automation.