A neural network–based synchronized computed torque controller for three degree-of-freedom planar parallel manipulators with uncertainties compensation

This study presents a new adaptive synchronized computed torque control algorithm based on neural networks for three degree-of-freedom planar parallel manipulators. The basic idea of the proposed control algorithm is to use the incorporation of cross-coupling errors of active joints with the adaptive computed torque control algorithm, online self-tuned neural networks, and error compensators. The key to the success of the proposed approach is to improve the trajectory tracking accuracy of the parallel manipulator’s end-effector while driving the synchronization errors among active joints to zero. The uncertainties of the control system such as modeling errors, frictional terms, and external disturbances are adaptively compensated online during the trajectory tracking of the parallel manipulator. Using the Lyapunov theory, it is proved that the tracking errors and error rates of the overall system asymptotically converge to zero. To demonstrate the effectiveness of the proposed control algorithm, compared simulations are conducted using MATLAB/Simulink [version 2013a] combined with Solidworks 2014.